Thoughts

Twelve Great Non-’Product’ and Non-’Data’ Podcasts for Product, Design, and Data People

Beyond the Data — Tue, 30 Jan 2024 13:00:00 +0000

In today's fast-paced world, staying up to date with the latest trends and insights is crucial for product, design, and analytics people. Ten years ago, when I would speak to audiences and ask how many listened to podcasts, there would be only a handful that would raise their hand. Now, podcasts have emerged in popularity as a valuable resource, offering in-depth discussions, expert interviews, and industry insights on various topics.

In this blog post, I am providing recommendations of twelve exceptional non-’product’ and non-’data’ podcasts that I recommend for product, design, and data people. These are all podcasts that I listen to regularly. These podcasts encourage my natural curiosity to continue to learn and grow. Whether you're seeking inspiration, practical tips, or thought-provoking discussions, these podcasts have you covered. Now to the podcast recommendations in alphabetical order:

a16z: The a16z podcast, hosted by Andreessen Horowitz (famed Venture Capital firm), delves into the world of technology, startups, and venture capital. This podcast features insightful interviews with industry experts, covering a wide range of topics such as product management, design, and entrepreneurship. Tune in to learn from the best in the business.

Akimbo: Seth Godin, renowned author and marketing expert, hosts Akimbo. This podcast explores the changing landscape of our culture, including topics like marketing, creativity, and leadership. With its thought-provoking episodes, Akimbo challenges conventional thinking and inspires listeners to make a difference in their respective fields.

The Brainy Business: The Brainy Business, hosted by Melina Palmer, combines behavioral economics and psychology to understand consumer behavior. This podcast explores the science behind decision-making, helping product, design, and data people create products and experiences that resonate with their target audience.

Conversations with Tyler: Economist Tyler Cowen hosts Conversations with Tyler, where he engages in enlightening conversations with a diverse range of guests. From economists to artists, this podcast explores various disciplines, providing valuable insights and perspectives for product managers and designers.

Econtalk: Econtalk, hosted by Russ Roberts, focuses on economics and its impact on our lives. Its tagline is “Conversation for the Curious,” so it's right up your alley. This podcast explores intriguing topics such as behavioral economics, decision-making, and the role of incentives. As a product manager or designer, understanding the economic forces at play can greatly inform your decision-making process.

Getting to Yes, And: Getting to Yes And, hosted by Second City’s Bob Kulhan, explores the world of improvisation and its applications in business and life. This podcast provides valuable insights into collaboration, creativity, and adaptive thinking, skills that are essential for product, design, and data people.

Killer Innovations: Killer Innovations, hosted by Phil McKinney, is dedicated to helping listeners become more innovative in their thinking. This podcast provides practical tips, strategies, and interviews with industry experts, guiding product managers and designers toward creating groundbreaking solutions.

The Knowledge Project: The Knowledge Project, hosted by Shane Parrish, dives into the minds of experts from various fields, including psychology, economics, and philosophy. This podcast explores timeless principles and mental models, offering valuable insights for product, design, and data people looking to broaden their knowledge.

More or Less: Behind the Statistics: More or Less: Behind the Statistics, hosted by Tim Harford, explores the stories behind the numbers we encounter in everyday life. This podcast delves into statistical claims, debunking myths and shedding light on the truth. For product, design, and data people, understanding how statistics are used and misused is essential in making informed decisions.

No Stupid Questions: No Stupid Questions, hosted by Angela Duckworth, is a podcast that explores the power of asking questions and the pursuit of knowledge. This podcast tackles thought-provoking topics, encouraging product, design, and data people to think critically and embrace curiosity.

The REWORK Podcast: The REWORK Podcast, by Basecamp, offers insights into productivity, teamwork, and the challenges faced by modern businesses. This podcast features interviews with entrepreneurs, managers, and thought leaders, providing valuable lessons and strategies for product, design, and data people.

The Vance Crowe Podcast: The Vance Crowe Podcast explores the intersection of communication, agriculture, and innovation. With a focus on storytelling, this podcast offers valuable perspectives on how to effectively communicate ideas and navigate complex industries. Product, design, and data people can gain valuable insights into effective communication strategies.

Hopefully, these twelve podcasts are as enjoyable for you as they are for me. If you have podcast recommendations, please send them my way. Note that in a future article, I plan to put together a more traditional product, design, and data podcasts.

Conclusion

These twelve podcasts provide a wealth of knowledge, inspiration, and practical advice for product, design, and data people. Whether you're seeking to enhance your skills, gain insights into industry trends, or simply looking for thought-provoking discussions, these podcasts will not disappoint. So, plug in your earphones and get ready to be inspired by the best in the business.

Happy listening!

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

Unlocking the Potential of Generative AI: 10 Prompt Engineering Recommendations

Beyond the Data — Tue, 16 Jan 2024 13:00:00 +0000

In the ever-evolving landscape of technology, Generative AI has emerged as a groundbreaking innovation, reshaping the way we approach problem-solving, creativity, and design. As product managers and designers, it's crucial to understand the concept of prompt engineering – the art of effectively communicating with AI models to elicit desired outcomes. This article delves into the essence of prompt engineering and provides best practices for harnessing the potential of Generative AI.

Prompt engineering is the process of crafting inputs (prompts) for AI models in a manner that maximizes the likelihood of achieving the desired output. It's akin to speaking a language that AI understands fluently, ensuring that your requests are interpreted accurately and effectively.

Prompt Engineering Best Practices: Prompt engineering expertise comes with practice, but here are starting recommendations to leverage to make your prompt engineering more effective:

Use clear language. Your prompt should be easy for the generative AI model to understand. Avoid using jargon or technical terms that the model may not be familiar with.
Be verbose, but break things into steps. Oftentimes, providing a more verbose prompt, as long as it is clear and understandable, will lead to a better result. For example, indicating “outline a blog post written in a professional manner to an audience of marketing professionals. The blog post is geared towards technology leaders and is driven around risk management and how to monitor potential impacts. This article should provide recommendations and risks to be aware about.” is a verbose prompt that should get better results than a more basic prompt “write a blog post to technology leaders on risk management.” When being verbose though, it also helps breaking your prompts into steps. This helps ensure there is clarity on the steps desired.
Define your AI assistant. Give the AI assistant as part of the prompt context on how it should respond. For example, maybe you are looking to have AI edit an article that you wrote. Provide as part of the prompt something like “You are an expert editor who has decades of experience providing critiques to product managers and designers on their writings. You do not suggest things that you are uncertain about.” this might provide improved output. For some models, you can update the ‘system prompt’, which provides the context of the model as a precursor and where you would typically define your assistant.
Define your audience. When it is relevant that you are doing research or writing to a specific audience (e.g., product managers), then provide the AI with the context that your audience is product managers.
Define your goals and objectives. Providing a generative AI model your goal or objective in a clear manner. For example, if asking a generative AI model to write a first draft, you may provide a goal of the style of writing (e.g., professional), you may provide a desired format, and you may provide a desired word count.
Be specific. The more specific you are with your prompt, the more likely you are to get the results you want. For example, instead of saying, "Write a story about a character who goes on an adventure," you could say, "Write a story about a character who goes on an adventure to find a magical treasure."
Provide examples. One way to enhance your AI models is by providing an example or two of what to expect. This will help provide the AI model context on the problem and format asked.
Give the model some creative freedom. Don't be afraid to let the model go off on its own a little bit. Sometimes, the best results come from unexpected places. Many models have a temperature setting where the lowest temperature is a deterministic model where the most probable output is repeatedly provided back, and the highest temperature setting is a highly stochastic model where you get a variety of results.
Understand the model's limitations. Familiarize yourself with the strengths and limitations of the AI model you are using. This knowledge will help in setting realistic expectations and crafting effective prompts. Models change rapidly, so don’t get stuck in one model. Be flexible by trying different AI models for different tasks.
Test your prompts. Once you've written a prompt, test it out with the generative AI model to see what results you get. This will help you fine-tune your prompts and get the best possible results.

If you are looking to learn more about prompt engineering, then there are many resources out there. One course that I have taken is Jules White, Ph.D. of Vanderbilt’s course titled “Prompt Engineering for ChatGPT,” which is available on Coursera. Of course, you could also try out asking different AI models about best practices around prompt engineering and see what they provide. In fact, when starting this blog post, that is exactly what I did with a few different AI models, and then I combined and edited what is posted here today.

Prompt engineering is more than a technical skill; it's a creative and strategic tool that can unlock the vast capabilities of Generative AI. By mastering this art, product managers and designers can revolutionize their approach to problem-solving, innovation, and design. Remember, effective communication with AI not only enhances efficiency but also opens doors to unparalleled creative possibilities. With a little practice, you'll be able to create effective prompts that will help you get the most out of your generative AI model.

Happy prompting!

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

Integrating Generative AI: Questions to Answer for an Existing Product

Beyond the Data — Tue, 09 Jan 2024 13:29:00 +0000

Generative Artificial intelligence (Generative AI), like ChatGPT, is the hammer that nearly every product person is asked to swing looking for nails as of the writing of this article. There are good reasons for the interest in Generative AI. Still, this post seeks to break things down from the perspective of a product leader or manager of one or more existing products on how to approach and evaluate Generative AI for those existing products. In a future post, there will be a discussion about leveraging Generative AI from the new product development perspective.

As a product leader, manager, or owner who has one or more existing products and is looking to leverage ways to one-up the competition, Generative AI might be that leg up. Still, it can also be a waste of time and resources. Here are some questions you should ask and consider when evaluating whether Generative AI is the correct hammer for your product:

Is there an existing client need that might be fulfilled better or faster by leveraging Generative AI?

Don't be that hammer-hitting screws with AI. Make sure there are one or more customer pain points where you hypothesize Generative AI can add value and with better value overall than other options.

To be able to do this, you need to understand what Generative AI is good at and what it is not. This is a constantly evolving space, so whatever here may change, and not all Generative AI tools are equal. Some areas that Generative AI is good at include:

Summarizing information, such as synthesizing a paper, podcast, or video for key details and takeaways
Content creation, whether text, images, audio, video, etc.
Information extraction from documents, audio, and video
Data analysis and interpretation and the ability for users to ask questions related to their data
Writing code, including rapid prototype creation
Customer service and support in the form of text and audio chatbots
Compliance as a second set of eyes, whether it is document editing, code evaluation, or any other area where Generative AI has significant background knowledge

This is a small list of areas where Generative AI can play a key role in leveraging in your product, assuming that your customer has a pain point where the solution might be one of the items above.

Are there other ways to implement a solution for identified customer problems, and is Generative AI better?

Generally speaking, there are ways to accomplish the same objective without Generative AI. However, that may not always be the case. There are some other reasons why you may leverage Generative AI even when there are other options available, including:

Quicker to Market: Sometimes problems that can be solved by Generative AI can be solved by other solutions, but implementing Generative AI may be quicker. This can be a good reason to use Generative AI.
Lack of Talent or Resources for Other Solutions: Sometimes there are alternatives, but the talent doesn’t exist in-house to leverage it or isn’t available to implement it. This is again another good reason to use Generative AI.
Marketing Angle: Sometimes, it is simply a marketing angle of being able to indicate that you are using cutting-edge technology, and you know that clients want to be able to say that they are using products with cutting-edge technology. Yes, this is a great reason to use Generative AI.
Enhance Team Generative AI Capabilities: Sometimes teams will simply be looking to leverage Generative AI as part of a solution so the team becomes familiar with it, its capabilities, items to consider implementation and help you going forward in other cases that might not be so obvious. This reason is completely legitimate as long as the expectations around it are indeed this and the measurement of success of an effort ties in with these goals and objectives.

The key here is that even though there might be other options than Generative AI, there still may be good reasons to choose a Generative AI approach. Below, you will see a number of factors to consider around economics, performance, and more in additional questions. These will factor into the business case for Generative AI. One thing people forget when evaluating Generative AI is to consider its practicality in a production setting and how it fits in strategically within an organization and not just a prototype space.

Are you looking to implement a solution as a me-too or a differentiator and build a moat?

Many Generative AI use cases are not really being done out of a competitive advantage in that other companies cannot recreate the solution. Instead, they often are done as part of a broader product offering that meets a series of clients’ needs. This is completely legitimate and all right.

However, if the strategy is to leverage Generative AI in a way of building a moat, then there are some key things to consider:

Data moat as a differentiator: Just like AI generally, Generative AI differentiation is mostly based upon the data moat. What data do you have or have access to that is different from your competition? This includes the breadth, depth, and quality of the data you have access to. The more you can build up a data moat from your competition for meaningful data in your space, the easier it will be to differentiate Generative AI solutions.
Unique model as a differentiator: For nearly all companies, the uniqueness of their model is based upon their data and investment in talent and processing resources to build the models they use. Generally speaking, the differentiator of the model is really the underlying data, so while calling this out as a separate item for most companies, it really is restating the true differentiator is the data moat.
Speed to market as a differentiator: A team that knows how to quickly leverage and implement Generative AI in a segment may simply have speed as a differentiator in that they will be consistently faster to market with solutions. Tied with this speed to market is a reduced cost to get to market.

How does data privacy play a role in Generative AI solutions?

Data privacy is really important no matter the industry, but certainly, in industries like healthcare and finance, there is special attention. Accordingly, it is really important to understand the agreements and environments of partners and your agreements with those partners. Further, it is important to not use key identifying information of individuals in training models. The same concerns you have with data privacy and analytics generally should be the same in Generative AI. However, you should realize that some leaders may have some extra concerns here based on some negative stories that had come out originally as Generative AI came onto the market. Understanding this, you want to make sure to clearly communicate how you make sure a customer’s data is maintained and used safely and with the customer in mind all the time.

Does the economics of Generative AI make sense?

Generative AI tends to be expensive in its pricing model for proprietary solutions. Now that pricing is coming down, my prediction is that this will not be discussed as an issue much in a couple of years from now, but for those looking to implement something in an existing product that has a significant number of users, then this is an important item to evaluate closely today. Note there are open-source alternatives that can be used, including some high-performing smaller models that may meet your needs and be an alternative to these pricey proprietary models.

Does the speed of Generative AI make sense?

Generative AI tends to respond in multiple seconds (sometimes 10’s of seconds) versus milliseconds. Speed is a factor of equipment and model usage, so Generative AI can be enhanced by leveraging higher-performance equipment or using models that are performant. This is an area of concern now still and something to consider, but over the next couple of years, it is my belief this will become a non-issue for most Generative AI use cases as models and equipment will be faster.

How important is Generative AI response consistency?

Generative AI tends to respond in a stochastic manner where different responses will be given for the same item. There are ways to counteract this, some leveraging the ‘temperature’ of models to make it a deterministic model, but that can also take away some of the perceived benefits of generative AI. I mention this item for awareness, but I really do not think this should be a concern for product leaders and managers.

How big of a concern are hallucinations for your use case?

Generative AI ‘hallucinations’ represent when generative AI simply makes up things that are not true. There is a recent classic case in which some lawyers were leveraging generative AI to write a court response that the generative AI used simply made-up case law and cited it in their brief. See Reuters. This is obviously a massive no-no for the use case for those lawyers, and now they face disciplinary action.

Before getting too concerned here, it is important to know that there are some good ways to counteract hallucinations. Implementing system prompts that require only having generative AI response when it absolutely knows a response, for example. Another approach is leveraging a secondary generative AI ‘auditor’ that filters anything being responded to as a truth filter is another approach. There are additional ways to counteract hallucinations, but even the raw models themselves are getting better each day in minimizing hallucinations. However, depending on your product and the use case you are seeking to leverage Generative AI within it, the possibility of hallucinations may be a greater or smaller concern and, in some cases, may completely eliminate potential use cases.

How important are product performance and uptime?

One challenge Generative AI has is that it is currently slower than a lot of other software performance metrics that people are used to in products. Instead of 10’s of milliseconds to provide a user response, it may be seconds or 10’s of seconds. Depending on the use case, this may be all right, but it is something to be aware of. The benefit is that more and more smaller models exist, and these smaller models are faster with the same computer hardware and can help improve speed. My view is that in a year or two, this will be a de minimis issue, but for now, if going live in an existing product, make sure to be aware of and test.

Related to performance is uptime. Our customers expect our products to work. Some Generative AI proprietary models have had some mixed performance in uptime history. Or even when up have had some significant degradation in performance. If leveraging a proprietary solution, then it is important to look into the performance and uptime consistency historically.

What are the success metrics for any Generative AI initiative into an existing product?

Making sure upfront how to measure value and success is important. Define that early, along with identifying how you will measure this when put in place, will make sure that you not only communicate to leadership the value provided but also make it easier for the next time you’re seeking to invest resources into enhancing existing products with Generative AI or create new products where Generative AI is a key component.

Conclusion

Leveraging Generative AI in your product can be a great way to improve the user experience, increase customer engagement, and drive business growth. However, it's important to carefully consider all of the factors involved before making the decision to invest in Generative AI. By following the steps outlined in this blog post, you can increase your chances of success when leveraging Generative AI in your product.

Happy experimenting!

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

Harnessing the Power of AI in Product Development

Beyond the Data — Thu, 04 Jan 2024 01:51:11 +0000

As the world becomes increasingly digital, artificial intelligence (AI) is playing a more and more important role in product development. Product managers and designers who want to stay ahead of the curve need to understand how to leverage AI to be more effective in their roles.

How can AI empower you and the product development process?

AI can help product managers and designers be more efficient and effective in a number of ways. For example, AI can be used to:

Analyze product data: Product analytics data can be a gold mine, but it takes work to mine. Luckily, AI can be leveraged in this process to get you going. If you are lucky enough to have a good data or product analyst, then this will make them more efficient. Sometimes, this AI is built into existing tools, but sometimes, you might combine an existing data tool to leverage a third-party AI framework for these insights.
Classify customer feedback: Understanding qualitative feedback through text responses in surveys or through a customer feedback mechanism can benefit from leveraging AI. AI can easily categorize feedback into segments and assign positivity/negativity levels.
Draft and edit content: There is a lot of content product people develop, ranging from product documentation, marketing materials, strategy overviews, survey questions, etc. AI is great at providing a good first draft of many types of content when provided with good prompting. AI is also great at editing content when prompted with this mindset.
Create visuals: One impressive and fun area of AI is leveraging it to create visuals. Take any idea and create static images or even videos based on prompts. Images and videos are often helpful in communicating stories to capture internal and external customers’ input but are oftentimes not used as much as desired because of the cost or resources needed to have good visuals and video.
Improve decision-making: AI can be used to provide product managers with insights into customer behavior and trends. This information can help product managers make better decisions about product development and marketing.
Market research: AI can help streamline the market research process, whether through looking through secondary research and summarizing key findings or formulating surveys to leverage in primary research.
Be a second set of eyes: AI is a great second set of eyes, whether it is reviewing content you wrote, evaluating the quality of a presentation, identifying issues in a piece of software, and much more. There are many ways that AI can be a second set of eyes that can help product managers avoid costly mistakes.
Roadmapping assistance: Leveraging various customer inputs and market research and identifying resources can allow you to leverage AI to help create and update your roadmap. More and more traditional roadmapping tools are looking at embedding AI directly into their tools to help product managers here.
Generate new ideas: AI can be used to generate new ideas for products and features. This can help product managers stay ahead of the competition and deliver innovative products that meet the needs of their customers.
Improve product quality: AI can be used to test products and identify bugs. This can help ensure that products are high quality and meet the expectations of their customers.
Improve customer experience: AI can be used to personalize the customer experience by providing users with tailored recommendations and offers. This can help product managers build stronger relationships with their customers and increase customer satisfaction.
Generate prototypes: There are a variety of AI-powered prototyping tools. This allows for product managers and designers to quickly generate designs and prototypes to help get customer input.
AI-powered products: More and more products are leveraging AI as a core part of their product offering. Developments in AI have made this integration process faster and at a lower cost. Everything from classification, summarization, image and video generation, document decoding, and much more can now be done more rapidly, more cheaply, and more effectively than just a few years ago. Product managers and designers are looking for ways to leverage AI and its many uses to solve customer problems.

There are many other ways that product managers and designers are leveraging AI, but this list will help spur your thought process of how AI can be a useful tool for you. Be on the lookout for more in-depth articles in the coming months here, diving deeply into how AI can be used in these specific areas and diving into some specific use cases and how to go about leveraging AI for them.

How to get started with AI?

If you're a product manager or designer who is interested in leveraging AI, there are a few things you can do to get started. First, you need to learn about the different types of AI and how they can be used in product development. There are a number of resources available online that can help you with this. Here are a few of the resources I have personally leveraged:

Podcasts:

This Week in Startups - AI Demo Edition (Website, YouTube)
Voicebot + Synthedia (Website, YouTube)

Online Courses (ordered in recommended progression):

There are many other courses, podcasts, websites, and more that I leverage to expand my knowledge of AI. My most meaningful learning has simply been getting my hands dirty through the use of different AI, which is oftentimes free or relatively low-cost. As of the time of this article, I have Bard, ChatGPT, Claude 2, and You.com all populate when my main browser is opened. Further, on my phone, I have a folder that holds ChatGPT, Bing, Microsoft Copilot, Poe, and You.com apps. Every day, I use different AI options to help me save time and perform better. Sometimes I will try the same activity leveraging different models to experiment.

Great product managers and designers are curious, and my guess if you are reading this is you might be above average in curiosity. Leverage this curiosity around learning by doing with AI, and you will be the beneficiary.

It's important to remember that AI is a tool, and like any tool, it can be used for good or for evil. It's important to use AI ethically and responsibly. When used correctly, AI can be a powerful tool that can help product managers be more efficient and effective.

Now, take the opportunity today to leverage AI in something you do. Remember to be on the lookout for more articles that will dive deeper into the topics identified above. Some of the upcoming articles will be about thinking strategically in leveraging AI in your product, possible ways to differentiate your product with AI, prompt engineering in product development, and more.

Happy experimenting!

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

Product Roadmap Success Practices

Beyond the Data — Thu, 14 Dec 2023 20:17:00 +0000

A product roadmap is a living document that outlines the vision, goals, and strategy for a product. It is used to communicate with stakeholders and keep everyone on the same page. A well-crafted product roadmap can help to improve alignment, focus, and execution.

Here are some best practices for creating a product roadmap:

Start with the customer. What are the needs of your customers? What are they trying to achieve? Once you understand your customers, you can start to develop a product roadmap that addresses their needs.
Set clear goals and objectives. What do you want to achieve with your product? What are the key milestones you need to reach? Your product roadmap should be aligned with your customer needs and overall business strengths and strategy.
Break down your roadmap into phases. A product roadmap is not a one-size-fits-all document. It should be tailored to the specific needs of your product. Break your roadmap down into phases. Focus on one phase at a time. The closer the timeframe, the clearer the phase will be displayed.
Be realistic. Don't try to cram too much into your product roadmap. It's better to focus on a few key features and deliver them well than to try to do too much and end up with a product that is not fully baked.
Communicate your roadmap with stakeholders. Your product roadmap is a valuable tool for communication. Use it to keep stakeholders informed about your progress and to get their feedback. This includes internal and external stakeholders.
Continually update. A roadmap is a living document that will be updated. If you have new information and/or strategy, then make new choices and make roadmap changes. Ideally, this update process is regular and done quarterly, but it could be more frequent.

By following these best practices, you can create a product roadmap that will help you achieve your business goals. However, don’t think the same roadmap will fit all organizations and audiences.

Here are some additional tips for creating a product roadmap:

Use a visual format. A visual roadmap can be easier to understand and follow than a text-based roadmap.
Use clear and concise language. Avoid jargon and technical terms that your stakeholders may not understand.
Be flexible. Your product roadmap is not set in stone. It should be updated as needed to reflect changes in the market or your business goals.
Use a system that makes it easy to update. There are different roadmapping software out there to assist here, but I am not making a specific recommendation. With that said, the majority of roadmaps out there still use things like Excel/Sheet and/or PowerPoint/Slides.

A well-crafted product roadmap can be a valuable tool for product managers, developers, and stakeholders. It can help to improve alignment, focus, and execution, and it can help you to achieve your business goals. If just starting this process out at an organization or at an organization with a challenged roadmapping process, then sometimes it is good to partner with someone with experience to assist in this process.

Happy roadmapping!

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

My New Book! Data Storytelling and Translation: Bridging the Gap from Numbers to Narratives is available!

Beyond the Data — Tue, 07 Nov 2023 14:00:00 +0000

In the digital age, data is the new currency. However, amassing heaps of data means nothing if it doesn't lead to actionable insights. It's not enough to present numbers; data needs a narrative to truly resonate with an audience.

Data Storytelling and Translation bridges the chasm between numbers and narratives. Learn the intricacies of translating raw data into compelling stories that captivate, inform, and inspire action. The book covers proven frameworks for converting data into compelling narratives, strategies to tailor data stories to different audiences, techniques to avoid common pitfalls and biases in data representation, the balance between aesthetics and accuracy in data visualization, and uses real-world case studies illustrating the power of effective data storytelling.

Whether you're a data scientist, business analyst, student, or decision-maker, this book offers the tools to articulate the true value of your data.

Note to Educators: If you are interested in leveraging this book in the classroom (college, university, or even high school), let me know. I’m interested in making that a success for you and looking to offer an optional virtual session for your students for the first of your classes that adopts leveraging this book as part of the curriculum. This session can be recorded and reused for future sessions. Remember, some sample syllabi are available for this book, along with practical exercises throughout the book.

Get it now

Why Data Storytelling and Translation Matters

In today's data-driven world, it's more important than ever to be able to tell a compelling data story and translate complex information and decisions among people. With so much information available, it's easy for audiences to get lost in the numbers. A well-told data story that properly translates information for the audience can help you cut through the noise and make your data more accessible and understandable.

Data storytelling and translation can also help you build trust and credibility with your audience. When you can clearly and concisely explain the meaning of your data, you're more likely to be seen as a credible source of information. This can be a valuable asset in any business or organization.

Chapters in the Data Storytelling and Translation Book Include:

The 10 engaging chapters that will help you be a better data storyteller and translator include:

Introduction - The Age of the Data Communicator
All Decisions Start with People
Start With Good Questions and Great Listening
Being Fluent in the Language of Data
Identify, Understand, and Frame Problems
Simplifying Insights and Through Metrics and Objectives
Painting Your Data Story
Leveraging Visuals to Share Insights and Compel Action
Leveraging Dashboards in Your Communication
Communicating Your Data Story

Get Your Copy of Data Storytelling and Translation

If you're ready to learn more about data storytelling and translation, I encourage you to get your copy of Data Storytelling and Translation today. The book is available on Amazon (paperback and Kindle) or at most major book stores.

Thanks for reading!

- Dave Mathias

Follow me on LinkedIn

Get your copy of Data Story and Translation today

Get it Now

We can all benefit from a little empathy and a map

Beyond the Data — Wed, 02 Jun 2021 18:50:32 +0000

Understanding your customers, your audience, and even people close to you is not easy. It takes effort. It takes intentional listening. It takes intentional action. One technique to help you understand others is through empathy mapping. It is something that I use a lot, and it works.

Empathy mapping is popular in the design world. In fact, when I Googled “empathy map” today, I got back 135,000+ results. There is no one way to do empathy mapping, just like there is no one way to understand your customer or audience. But if you haven’t, then give empathy mapping a try.

Let’s start with getting into what is empathy mapping and how to do it, then we will dive into the reasons why empathy mapping is valuable, and lastly we will finish off what do you do next from with your empathy maps.

Why do empathy mapping?

There are five reasons I like using empathy mapping:

Intentional focus on defining customer segments. Putting together an empathy map first requires you to focus on segmenting your customer or audience into segments. Identifying customer segments is something that people in marketing, sales, and product, for example, do all the time. But, really segmenting our customers (whether internal or external and oftentimes not just by department) is not something many of us do but should do.
Intentional focus on understanding customer segments. Once you have your customer segments, you need to understand what your customer “does,” “says,” “thinks,” and “feels.” This is not easy. It means we need not only to listen but translate information that is not always directly said and without putting our own bias on it.
Forces me to question my assumptions and understandings. We all tend to be more confident than we should be. I know I suffer from this. Therefore, the way I do empathy mapping forces me to identify confidence in my understanding and data sources for them. For example, all too often, our understanding of our customers is because of the loudest voice only. Identifying my understanding of the customer as only the loudest voice will push me to get other voices.
Team-based understanding: Empathy mapping is something that is best done with your team or group. There are certainly individual components of the process, and it can be done entirely by yourself. But, in my experience, the best empathy maps are generated by teams together.
Good ongoing reference material. The empathy mapping process is great, but even more valuable is revisiting the resulting empathy maps as an ongoing reference. Sometimes, these empathy maps will be generated into “customer personas,” but sometimes, it just might be the empathy map itself. They are useful to revisit yourself, like when you are putting together a big communication to a customer segment. But they are just as useful for sharing and transferring knowledge to new team members.

Each of the reasons above by itself is, in my opinion, valuable enough to leverage empathy mapping. However, all five reasons together make empathy mapping an important tool in my playbook.

How do you do empathy mapping?

There is not a one-size-fits-all way to do empathy mapping. I like to keep it simple but with a twist.

Identify customer group and context. The first step is identifying your different customer groups or segments. Each customer group or segment will be a separate empathy map. For example, if you are a recruiter in human resources, a customer segment might be technology managers. Further, you identify the context or interaction that you are seeking to map out. Again, taking the same recruiter example, an interaction with your technology manager customer might be a new technical employee recruiting and hiring process.

Now that you have a customer group and context established, you need to map what that group “says,” “does,” “thinks,” and “feels.” The customer segment and context are in the middle. Then, there are four quadrants laid out for you to complete: the says, does, thinks, and feels. Oftentimes, a physical or virtual whiteboard is the place to do this.

A blank empathy map template is on the left. On the right, is an example first draft template for our recruiter working with her customer a technology hiring manager.

What information do you have to have to support what you enter? How confident are you in what you identify? Certainly, if you are an experienced recruiter working with technology managers then some of this may seem intuitive at first. However, what other information mechanisms do you have as input? Is there customer survey data? Is there customer observational data? Is there customer’s customer survey data? It is important to not only write down what you think you know but how confident you are and even how do you know it.

There are two ways that I indicate confidence. Sometimes using different colored sticky notes can represent different levels of confidence. Another thing I do and have been doing more often is using an open circle representing low confidence (e.g. maybe have some anecdotal examples or data); a half-filled circle as middle confidence (e.g. have some survey data or lots of examples); a filled in circle when I have high confidence (e.g. have lots of supporting data and examples that are consistent). Most things will be low or middle confidence in my experience and that is alright.

The first time you do an empathy map, you will generally have a lot to write down in the “says” and “does” sections. However, it will take longer and be more sparse in the “thinks” and “feels” sections. This is alright. Even when completing an empathy map, you should go about doing it

The best empathy maps come from a process of doing it as a small team or group. The participants should work with or interact with the customer segment, given the context being mapped. Doing an empathy map as a team leads to a better end product. Additionally, the process of doing empathy maps can be a good team builder. Here are a few suggestions when doing empathy mapping as a team or group:

Build in individual processing time at the beginning of the empathy mapping process. It will help ensure one or two voices don’t just dominate and that both extroverts and introverts provide important input.
Everyone should provide input and support it with reasoning or data. It is alright to have a gut feeling or reasoning identified but then identified as low confidence is appropriate.
The empathy mapping process should be done in a build-and-then-prune mentality. First, build the empathy map with everyone’s contributions and no debate or criticism for contributions. Then, in the pruning stage, you question or refine after everything is initially built.
The empathy mapping process is iterative, and groups may need to go back and get more information intentionally.
Build a feedback loop so that you get input from some of those in the target group. This input should be sought while providing your customers an admission that you seek to understand them better and value them. Hence, it is why you have invested in this empathy mapping process.
Empathy maps should be built with a built-in revisiting time. Things change over time and need refreshing. Identifying this need and timing in advance helps make sure the most value comes out of the process.

What to do with your empathy map?

The empathy mapping process itself is certainly valuable. When done by yourself, it will help you intentionally define, understand, and refine your understanding of customers and context. When done with a team, it will provide a better shared understanding of the customer and a context along with a team-building activity.

Taking these understandings and applying them to your day-to-day interactions with your customers is important. I like to ask myself two simple questions post-empathy map:

What understanding is different from what I had before?
With that different understanding, what should I do differently?

For example, this may mean I communicate differently with my customers.

Empathy maps should also not be hidden away. A new team member who joins a team should have empathy maps shared with her. Sharing it with her results in a better understanding of your customer quickly and more team alignment.

Eventually, you may translate your empathy map into full-blown customer personas that are an extension of your empathy map. In a future post, I will write about developing and leveraging customer personas. However, in short, customer personas are fictional representatives of a customer segment that can be referenced and communicated about.

In the end, understanding our customers is vital to our success no matter what we do. This is even true in our personal lives. Empathy mapping is a tool for your toolbox to be better at understanding your customers. I encourage you to give empathy mapping a try and see how it works for you.

Happy customer understanding and leveraging empathy in your life!

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

Ep 44 - Haydee Hernandez - Building a culture of experimentation

Beyond the Data — Fri, 30 Apr 2021 12:54:33 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

Haydee Hernandez, Director of Product Management @ CollegeAdvisor.com

This was an episode where I talked with Haydee Hernandez. Haydee and I met at ProductCamp Chicago more than 5 years ago. Like me Haydee is an avid believer in constant experimentation to make the best products and experiences, and leveraging data from those experiments to make decisions. Listen to Haydee and my conversation around building a culture of experimentation.

Haydee’s BIO: Haydee is a product leader who melds a strategic mindset with hands-on tactical preparation to deliver maximum value. She is able to craft a distinct product vision through the synthesis of both quantitative and qualitative data. Her work setting up A/B testing frameworks and processes accelerates learning and growth cycles within large and small organizations. She has overseen million-dollar integration efforts, launched with minimal customer or organizational disruption. Her ability to collaborate with executive leadership and cross-functional teams leads to organizational alignment and enthusiasm. By deciding to optimize or slash projects based on outcomes, Haydee proves she is a person able to take ownership of the results. Haydee brings nearly 20 years of experience launching digital product initiatives in B2C and B2B companies, agencies and consultancies.

Connect with Haydee:

LinkedIn: in/hernandezhaydee
Twitter @haydee
Current Company: CollegeAdvisor.com

go Beyond the Data Ep 44 – Haydee Hernandez

Haydee Hernandez, Director of Product Management @ CollegeAdvisor.com

in/hernandezhaydee

Machine Generated Transcript via Descript

Remember to subscribe to the go Beyond the Data Newsletter where I share data-driven and human-centered insights from others.

Have a great day! Dave

Dave Mathias: [00:00:00] Hey everyone. Good to have you on another episode of go Beyond the Data with me today is a friend of mine, Haydee Hernandez out of Chicago, but soon to be Miami .

Haydee Hernandez: [00:00:09] Absolutely. Thanks for having me on Dave.

Dave Mathias: [00:00:12] Great to have you here. We actually met at a Product Camp. I don't know if it was like 2015 or 14 or something like that in Chicago some years ago. We actually did a session around experimentation and building a culture of experimentation last year.

And thought it'd be great to have you here today. And let's just talk about experimenting a little bit more and have a little fun. You have an interesting background, we always have good conversations, so good to have you.

Haydee Hernandez: [00:00:36] Thanks. Dave, what do you think your viewers would like to learn more about?

Dave Mathias: [00:00:40] I think you have that data mindset, but you also have, you're an anthropology undergrad, right?

Haydee Hernandez: [00:00:46] Yes. I was an anthropology undergrad. A background that you wouldn't think would make much sense with technology, but actually has probably been one of the most useful skill sets I've had.

Dave Mathias: [00:00:57] Well, especially as the more important aspect of user experience and product design and how we really are trying to understand the users better, our customers better. I think anthropology, which is great. I see somebody with that kind of background I get excited. So I know it may be atypical. I is, but it's great to see, but then you did a masters and it's funny because you did the anthropology on the East coast, right? Like I think in the DC area at a, was at George Washington you went to university.

Haydee Hernandez: [00:01:23] Yep. That's where I went.

Dave Mathias: [00:01:24] And then you went to the West coast, you went to Berkeley, UC Berkeley for your MIS. And so what got you from anthropology to say, okay, I'm going from anthropology, but then I want to, the MIS the masters in information management systems.

Haydee Hernandez: [00:01:39] It's I think a number of people who are starting off their careers and then they fall into something that is like magic. So when I graduated from GW, I had graduated with an anthropology degree, but I was actually working in a defense consulting firm. And so upon graduation, the firm got all these contracts and really needed to explode in terms of the number of resources that they had.

And yeah. They asked me to stay on in a full-time capacity in HR. But the condition was is that I build out all their HR needs, but without having a lot of other resources and back then over 20 years ago, that meant computers. So they were willing to throw computers, programs, training at me.

And so next thing I'm an anthropology graduate, but I'm building an intranet, designing flat file databases, then moving to relational databases then learning publishing software. And after a while, it didn't take me long to realize I really love this technology piece way more than the HR piece.

And what can I do with that? And so I discovered Berkeley was kicking off a new department and they were approaching technology from this interdisciplinary viewpoint where it was a combination of business meets computer science meets social science, which really resonated with my anthropology background.

And I was interested in leveraging that research. Buddy that you did from anthropology into how do you use that to understand how people can use technology for their needs versus having technology force people to adapt to how technology is created. And so Berkeley really opened the door for me to explore that area.

And next thing you know, I wind up graduating and I'm in the tech space.

Dave Mathias: [00:03:26] That's a great story. And it explains a lot as the path that you've taken after it, because even in your career you've hit on a lot of things. And like you said, you went from HR, but then you became an analyst after Berkeley eventually, and then you went to user experience and then of course, you're now in product management, you're a director of product management and we'll get into some of the things you're doing now. But one of the themes that I know we've chatted about a lot is around experimentation and the need to, learn quickly and leverage whatever's there. And certainly sometimes that's a technology and certain times it's in the anthropology world it's observing and having conversations with people and just people to people like, which of course this is COVID now. So maybe our skills are not as good as they once were.

But can you talk a little bit about like, how your journey progressed to get you now as a director of product at a pretty interesting startup for?

Haydee Hernandez: [00:04:16] Yeah. The career journey over the last 20 years is actually been pretty exciting because as you've noted, I've been in a number of different roles all within the tech space.

I've been an analyst, I've been a project manager. I spent quite a bit of time doing information architecture as well as some user research. And then most recently in the past 11 years leveraged all of that to really look at product management holistically, and to take all of those pieces and really focus on, being very user-focused then in the type of work that I do not making assumptions on what is the right solution.

I'm looking to experiment and ideally develop data to whether it's qualitative or quantitative that can substantiate the direction that we're going. And so it's a, it's been an exciting time. Each of the changes that I've made has. Been more towards trying to push myself early in, earlier in the life cycle.

So that, as an analyst, you might be coming in later, once everything has already been decided what you're going to be doing. Same thing with being an information architect or doing project management, but on the product side, you're at the forefront, you get to be at the table when conversations are being had about what is the strategy for product. How does that relate to the overall company strategy and their bottom line goals? And how do you get from, to close that gap between what the goal is and where you're currently at?

Dave Mathias: [00:05:35] Yeah and so certainly as part of that, of course is running many different experiments to help develop that strategy and to learn.

And the session we had in, I think it was last fall virtually it was around that building a culture of experimentation. And I know you, you've, obviously we experiment in many different ways and there's many different scenarios, and I'm sure you have hundreds of different stories, but maybe if there's a couple of stories that may help listeners understand better of some lessons around how if they wanted to build a culture of experimentation. What are maybe a scenario two that you went through in your career? A story that you could share with them and successes.

Haydee Hernandez: [00:06:09] Sure. I think the first thing is that I often hear concerns about, the level of difficulty and effort involved with just kicking off an experiment. If you are in an environment where that's not a common practice or maybe there's just also not very much in the way of resources, how do you get started?

And in my current role one of the things that was exciting is that at a startup you get to play a lot of different roles. But you don't have as many resources as you do compare it to like when I was at like a Hyatt or Ticketmaster. And so you have to be pretty creative about how are you going to establish whether a feature is really viable or not. And so very recently I wound up launching a new feature within our portfolio, but the feature was something that, we didn't feature flag cause that would take more development time. It was quick and easy for us to spin up, but how to release it was really the question. And in our situation, what we wound up doing is actually I launched it as an email campaign where the link is publicly available if you know the exact link. However, if you don't know the exact link, you don't know that feature actually exists in the entire product. And so we, cherry picked who our target market was and we sent out the emails and then we just did very basic tracking for like the open rates for the email campaigns and what the click-through rate was.

And then beyond those basic email metrics leverage something like FullStory where you can actually watch user sessions to see how people are interacting. It also tells you what the average session length time was for the feature, which was over 12 minutes so we knew we already had a winner. The email campaign performed well, but the feature just really drew people into diving in deep and it captured their attention. And then when they came back and had even longer session lengths. And so it was very exciting that in a very short window of time, we could spin something up, check to see if this was a feature that was worth building out more. And establish, it's worth the money and the investment to not only develop the feature out more, but then to add it to the overall portfolio, because as many product managers know sometimes you, I have this issue of feature bloat where you have so many features and you're constantly maintaining them, but you don't really know if they're valuable or not.

And so this was a nice way to introduce something new without having to worry about everybody being invested in it yet, until we knew for sure we wanted to maintain it.

Dave Mathias: [00:08:36] Yeah, that's a great background to let the listeners in. Now, one thing I, of course I asked that question is even before you get to putting together the test How did you go about determining what is success and getting expectations on this and aligning things or how much effort was spent there or how much was okay, this is. Not going to be too hard to put out there and let's just see what it is. That's not set expectations. How much pre strategy work was put together.

Haydee Hernandez: [00:09:04] In this case, it was something where we knew that this didn't exist at all. And then I am actually representative of one of the users and we had also done a nice survey of several of our own internal people who were client facing and throughout the idea at them. And they, it resonated with them because they thought, Oh, we've heard the same complaint.

We've heard about the same type of problem from our own clients, but we just didn't know how to solve it. We did the best to be able to provide advice and guidance. But a tool like this could really help solve that problem on a larger level. And they hadn't seen anything in the market for it.

So this was one where. Between my own personal experience, being inspired by that checking in with others who do work with clients to see, does it resonate elsewhere? And then just getting just like a general backing internally, it was worth just throwing our hat into the ring because it only took a couple of weeks for us to really get the key pieces of data together enough for an MVP. The investment that we made felt small given the total potential payoff and how easy it was for everybody to just come to agreement that this was something that they really wanted to at least try out.

Dave Mathias: [00:10:12] Excellent. That's great background.

And certainly it sounds like you had your anthropologist hat on initially and leverage that. And then of course your more analysts type hat a couple of weeks later. So that's good to hear. So how is that different now, obviously you talked about a couple other places that you had worked at.

I think when we met, you were at Hyatt some years ago, and certainly that's a big organization an organization like that, and obviously a culture of experimentation. I know we've chatted. On some of your experiences there. Can you talk about that? Because obviously a business that's been around and successful for many years, many decades.

But it's a very different industries. So how do you In an environment like that, bring in experimentation and get people bought on board, especially your folks that may not be as, as used to that.

Haydee Hernandez: [00:10:51] That a challenge, because then you're working in the flip side where you're at a very mature company who has identified a model that works really well for them.

They've been successful, right? In the market and they're one of the primary competitors. And getting people bought into the idea of experimentation can be challenging because it's different than the everyday thing that you're doing. And if everything that you're doing each day, winds up.

Still moving the needle, then why change? But this is where it really comes down to like personal passion. And I was always interested in experimentation, loved the idea of user research. And when I was at CDW they invested heavily in user research there. So when I moved to Hyatt, that was one area that I thought, Oh, this is an opportunity for us to do more in. We all have projects. We all have do deadlines that are due. But they were other like-minded people and what it turns into is finding like-minded people who are willing to use part of their lunchtime or their free time, or stay a little bit later answer some questions here, what they would seeing inside of our analytics product answer some questions as to what they thought might be some interesting ways of tweaking our current design and then creating an opportunity for the managers to say yes, easily, because we had developed a very simple business case for a quick change that could be done relatively easily without development help. And that then from there, we just piggybacked on success after success so that we each small incremental change that you make at like a major company, like Hyatt, you can see significant gains on the bottom line.

And so that was something that over time they fought. Wow, this is worth investing in. But that comes from when you're willing to put in the passion, put in some of your own time and get other people bought in who might be peers who could help you there. There's not always a guarantee that it'll work, but, in my case it was very successful.

Dave Mathias: [00:12:49] I think a couple of things that you're saying out of that, which I think just to reiterate, because I think they're very valuable for our listeners is certainly start small, find your tribe and that tribe that's really passionate about the same thing . They believe in that and they're willing to spend a little extra time, spend some of those lunch hours and after work and before work and maybe even some of those weekends. There's people like that at every place, no matter how big of a company or how small of a company. And finding those folks and not trying to, try to put the tailwind in your favor as opposed to hit your headwinds. So what is the thing that you find the most enjoyable in the process? What sort of makes you thrive in that experimentation? What's the thing that you enjoy the most?

Haydee Hernandez: [00:13:27] For me, it's actually seeing what the results are. Especially when we can tell the story. You can get numbers that will show how people are behaving. So you may see, are they converting more? Are they getting through the process faster? Are they using the feature? But I also like to piggyback that with the qualitative side.

So whether it's like remote user sessions, focus groups, or one-to-one time being able to hear through their own voice as they're going through the process, what they're thinking. That opens up enormous doors, because then it provides a color to those numbers that you don't normally get to see. Sometimes you see a number and you may misinterpret why that number is the way it is.

And so when you throw on that qualitative side, you get that human aspect. And perhaps that's the anthropologist in me that I love talking to people and I like to see why they do what they do. But when I have that information and I have the quantitative data, what it does is it helps me improve every subsequent project because that nugget of gold for one project may have also unearthed, just like a small little, like rock that I could leverage to build in a different project. And so it's seeing that continuity across your users so that you have a stronger sense of what that persona is. Maybe identify personas that didn't previously exist and it just makes the whole work, very exciting. So that I liked seeing that combination of the two coming together.

Dave Mathias: [00:14:58] So for people consuming that story that you're putting together, because certainly you're good at being able to take, okay, the quantitative and the qualitative aspects and bring the information together. What's your process in trying to merge that together and relay stories to others? How do you go about that?

Haydee Hernandez: [00:15:14] It varies depending on what I have available. And normally I do the users to be able to speak for themselves. So in places like CDW or Hyatt, when we have recordings, being able to clip the recordings so that you can show your stakeholders exactly what was happening so they can see the struggle that the users have.

Or you can see how quick and easy something was, the joy that they had in doing something. And so that is huge because it really requires no effort on my part. It's genuine and authentic. Other times it's easier to combine in slides where you put the testimonials and you write out verbatims from what your users have said, and you pair that with key stats so that they can see both sides together.

It just kinda depends on what the most effective way is to tell that story. Sometimes the numbers are convincing and to themselves. So even if I have the stories, they may be things that I just speak to over the course of a presentation or share during some meeting. But I keep them all in my back pocket so that whatever is the most effective at that organization is what I try to move towards.

Dave Mathias: [00:16:22] Yeah, a you're in Chicago now, second city, of course, an improv. And so you're almost doing a little improv right in that storytelling aspect. And I think the best storytellers are able to they're prepared for different scenarios, but they're able to shift very quickly.

One of the things that thinking about is okay, obviously, When you have a lot of clicks, a lot of digital data coming in, you feel like you really understand your customer well, but oftentimes there's scenarios or different types of products or people that are not your customers.

How do you go about collecting data beyond just the click data? What are the ways that you like to gather data beyond just sorta like. The mail open rate or the click through on the website rate or these different types of digital data that we have a lot more coming, but we have our people that are not our customers now users that data, which we're not really seeing, but someone else is getting data, why they're using them.

And then also for folks that like, like you were at Hyatt and there's a lot of other things, of course, like you're using a hotel, your different formats so what's your other process of trying to gather data outside of just that straight digital data?

Haydee Hernandez: [00:17:26] There are two things that I wind up using one is focus groups.

So that's where you identify a group of people that are either like in the target market that you're already in, or maybe in an adjacent market that you want to move to and you identify those people. You invite them to somewhat informal session, right? You ask them a lot of questions to just start trying to understand what are the problems that they are facing.

How do they look at things? What are the things that they find easy? How do they go about solving those problems right now? And that I find is the greenfield space. So that's very exciting when you're working in an opportunity where you can define something that's never been defined before. So focus groups are good in that space.

When I was at Hyatt. Then there's also recruiting for people that may not be like regular Hyatt users. And so that's just going out, whether it's use a service identifying other individuals who might also fit the target market and then inviting them. With those, we did remote user sessions so that we could get a lot of feedback in a short period in time, because we could define a prototype or we could ask questions and then have them go through it, recorded on their computer and then in the morning we'd have a dozen or so to be able to look at, to walk through and hear. So we could scale really fast at hearing a large amount of feedback without having to invest a lot of our own time at night and make sure that we were also being consistent with each and every single user that we were generating feedback.

So you can do it at the individual level, or you can do it within a group.

Dave Mathias: [00:19:04] Excellent. That's some great input. obviously you've been in so many different scenarios and different roles that your experience is such a wealth of knowledge. But I think of if you're going to say the one thing or one of the main things that you did in your career, but at the time you might not have thought of how important it was. It could have been a class or it could have been a project you're on or anything like that. At the time you were probably not gun ho about it and not liking it. But at the same time you think you learned a lot from it that ended up really helping you throughout your career. Is there anything that you can think of that might fit that?

Haydee Hernandez: [00:19:36] Actually, it's funny, you mentioned that because very recently I reached out to one of my old professors at Berkeley. Her name is Marti Hurst. And so her specialty is actually within data visualization. And so she's done research for decades in that area. And I took a class with her and it was the most impactful class I've ever taken in my entire life. And I never would have expected one class to have been useful over and over again, but it has it really gave me a good grounding in the basics of user research.

And one of the articles that she actually suggested that we read and I said, I still have. In fact, since I'm moving, I actually found the article again. Just a couple of weeks ago, it's called prototyping for little fingers and the whole point of it was building low-fidelity prototypes. So getting yourself unblocked from whatever tool that you're using, you can use Miro, you can use Marvel.

They're all. There are all these cool apps. It pulls out their prototypes on the fly now. But this article was really about building physical prototypes like with cardboard or construction paper, crayons, markers and using that as a method, you engage with your user and I have to say, I.

Completely underestimated how much fun it was going through these exercises. Working with somebody after you designed a prototype on a piece of paper that they can literally move everything around. But what I found though was not only the value of prototyping in like the simplest quick, quickest way, which you can do that easily.

Paper, but that like that the tactile world is an exciting place for people to get energized and that you can draw so much more out of them and engage them in conversation more effectively. The defenses are down and it becomes more fun. And even though there's not as much prototyping in the physical world now, Especially with COVID.

It is still something that I try to figure out, how do I get that same level of energy and excitement? How do we engage people in a virtual world where they can share feedback, whether it's like stakeholders giving you feedback on what they want to see, or whether it's users giving you feedback? How do you get that from them in a way where it's fun, it's energizing and they feel open to make suggestions and they don't feel locked in with what you w with what you're presenting them.

Dave Mathias: [00:22:02] That is great feedback and I think you're spot on where it really levels of field a little bit. We all be kids again. And I think as you get older, you become more of an adult and you're like, we have more filters and more barriers. And we put up things in some of these are good, but some of these are probably not so good. And it really it's to really enhance innovation and creativity and breaking down those barriers being a kid again, this can help, asking those questions that we're more likely going to do in those situations. One of the things that we oftentimes have a hard time admitting, but we always say fail fast. Like these terms are putting out there, right? This, these embrace failure.

Like you hear this a lot but often times organizations don't really practice what they preach here. But even when they do I think learning, we learn from our failures. At least I personally feel like I've learned from failures a lot more than I feel like I've learned from successes. If there's a failure in your career or somewhere along the line that you're like, Hey, what did I learn from that? Is there anything that you could highlight that might be something that you think at the time you're like this is really a failure, but now I'm feeling like I learned so much from this and this is how I've used that.

Haydee Hernandez: [00:23:07] Actually failure can be a really interesting tool to educate stakeholders is what I found. Sometimes what is with experimentation that maybe certain stakeholders have ideas already in their heads about what they think will be very successful. And what they're convinced is worth investing large quantities of development time for and experimentation I find winds up being an effective way to be able to address that because one thing that you learn in doing experimentation is you're not always right. And so you always have to be willing to just wait and see what the outcome is. You can make your best guess you can listen to your gut. But at the end of the day, it's really about seeing what the end results are. And so I find what's interesting is using experimentation to then help understand what like stakeholders are thinking, like what they feel really strongly about and what they think is going to be super successful, breaking it down into some sort of quick MVP that we can test and then throwing that out there. And in some cases, when I've done that you find that. It's a total failure, like it didn't improve conversion. It is something that didn't wind up generating any improvement, like at all, it might just be a net neutral. And so those are really eye-opening because, even if you feel that this might not be successful, somebody else may feel very strongly.

Just like sometimes you feel something's going to be really successful in somebody who's questioning. Oh, that's probably not going to go anywhere. But leave it to the experiment to see. You know how it actually does do. And then I think biggest takeaway becomes that everybody one becomes more willing to accept the fact that they don't always know the right thing and that when people believe that, then they're much more willing to try these MVPs out first, before going and investing full blown into development.

Which is a better use of the development resources. It helps us really understand like where the real value is and if something is worth investing in. And so for me, failure is actually more of the recognition that I don't know what's right. And that each experiment is going to teach me something.

Dave Mathias: [00:25:04] Excellent. That's great feedback. I think just getting everyone on board with that as trying to build that culture of experimentation and that acceptance is we're trying to learn fast, right? We're not trying to fail fast. We're trying to learn fast and getting everyone on board and just, Hey, this is what we did. This is what we learned from it. And let's move on. If it conversion is a gap or whatever it is. You mentioned one thing with the I'll have to look at that prototyping for a little fingers. So thank you for that recommendation.

If somebody has interest in getting more into learning more to be better with experimentation. What would be a good thing that you would recommend?

Haydee Hernandez: [00:25:35] Actually, this may seem a bit off topic, but it's one of the areas that I'm exploring a lot right now is like within the area of productivity. And with experimentation, I find that one of the challenges when you're successful is just how many experiments you're running. And that it's tough to juggle.

And then also within product management, it's also tough to juggle. And so I have become incredibly fascinated with this gentleman named Tiago Forte. He does a productivity blog. He also does a number of really. Fantastic. YouTube videos about building a second brain. And so how do you go about capturing all the data that's coming at you, which is massive whether it's personal or work and then how do you go about distilling that data so that you can go deeper in?

And so he has this really interesting methodology for developing the second brain and then leveraging it in order to be more productive. And so for me where I'm at in my career, that seems to be more of what's resonating is I want to be able to do more with what I've learned. I want to be able to access it faster.

I want to be able to find ways to make it more accessible for others, because as I learn more stuff, it's in my head, but I needed to be in our community's head, whether it's at my company. Or across other people that I meet in the product management world. And that's what I've been looking at right now in order to try to help me level up based on where I'm at.

Dave Mathias: [00:27:05] Excellent. Is there any technology that related to that productivity that you typically use or have been trying that you've been excited about?

Haydee Hernandez: [00:27:13] I've actually been really excited about Notion. I love databases, but it's like database on crack because I can create cool databases, link them to one another, but make them look like web pages.

And so I get that feeling like I'm creating my own personal internet, but yet it's a database at the same time. And so I find that it's been incredibly helpful for keeping track of all the books or articles that I want to read. Being able to be a storehouse for all the comments and notes that I've taken for Kindle or on Instapaper.

And it's just like a great source for me to keep track of quotes projects that I'm working on general areas in my life. And so it's it's so much fun. I'm actually using it at the office too, to help use that as like our internal Wiki as well, because it is so quick and easy for people to learn.

And it turns out that it's it's gotten a lot of more popularity lately. Several people that I've introduced it to have just jumped on it, like a bandwagon as well. It's what I do to basically how's my second brand that I'm building.

Dave Mathias: [00:28:21] It is it's funny. Cause I was literally going to say like Notion, but I was like, no, I don't want to like prime her at all. Like right before we got on what did I do, I pulled up my Notion where I had some notes based on before this. I totally hear you. And there's other pro I used to be a big Evernote person, like I, a premium Evernote person. But then I switched Notion and just, there's a lot of if you are somebody that's trying to find some more productivity in your life. I still feel like there's so many features in it that I still don't understand. Like I'm still like learning things all the time. So in fact, every so often, go on YouTube ads or check out the gentleman that you said, cause he probably has some videos around there. Cause trying to learn more about I'm sure I'm not using all the potential that even it has.

Haydee Hernandez: [00:29:01] Yes. Tiago Forte is great. And then Ali Abdaal I may have massacred his last name. But he is also now serving as a mentor with Tiago's classes. And I love his his YouTube videos as well, because they're good. Not only in terms of building the second brain, but also just in general productivity tips. Because it is so hard nowadays with everything that's being thrown at you, how to juggle.

Dave Mathias: [00:29:24] I totally hear you there. It's been great to have you on today. Is there anything that you before we go that you're excited about, that you think you could share with the audience on anything that you're maybe working on now, or you've seen lately?

Haydee Hernandez: [00:29:35] I'm just really excited about the startup that I'm at right now. CollegeAdvisor.com is a fantastic place. We've been scaling up really fast and it's it's a place that I just really deeply love. Not only because as the head of product I see a tremendous amount of opportunity for growth, but my own daughter has actually gone through and use the services. And so it's incredible to see that in the time of the pandemic, when she's needed helping guidance. And it's hard to do that with the lack of guidance counselors, without teachers being so readily accessible that she's had that one-to-one coaching. And yeah. Now she's off to USC with a full tuition scholarship.

And so it's incredible. So I want to be able to give more of that type of experience to other parents and to help students who are facing similar troubles, where it's just very challenging for them to imagine how they're going to go through the admissions process, even with us, having vaccinations.

And for me, that's my next really big adventure is just having a lot of fun at at the startup and seeing it grow.

Dave Mathias: [00:30:33] Yeah, that's exciting. I know we were chatting about that earlier and that you're moving in the same month your daughter's moving out, out to USC. You're going to be going down to Miami. She's going to LA. So it'll be a lot of those cross-country trips. You'll put, you'll build up some frequent flyers miles when we're back flying a little bit more. It's great to talk with you wish we were in person, but we'll the next time I have to grab this and maybe a coffee shop on South beach somewhere and have a conversation out there.

Great to talk to the hand Haydee and have a great rest of the weekend.

Haydee Hernandez: [00:30:59] Thanks dave, talk to you soon.

Dave Mathias: [00:31:01] And one other thing I should, of course I forgot to say is how would people get ahold of you if they want to learn more about you, what you're doing or they want to get their kid, a scholarship and move down that process. Is LinkedIn the best or where's the best way to get ahold of you?

Haydee Hernandez: [00:31:14] Absolutely. Reaching me through LinkedIn is an easy way. I regularly check that. So drop me a note. Let me know if you need anything, whether it's with admissions, experimentation, or if you're just looking to move into a product management career, I've helped mentor people before in that space as well.

Dave Mathias: [00:31:31] And the company you're with now

Haydee Hernandez: [00:31:32] CollegeAdvisor.com.

Dave Mathias: [00:31:34] And we all could use a full ride looking back from either the parents for their kids. So I'm sure people will be reaching out to you.

And I was like, Hey, how did you go about doing that? So congrats again. That's awesome. And have a good rest of weekend.

Haydee Hernandez: Thanks Dave. Thanks.

Dave Mathias: Cheers. Bye.

Learn more

Permalink

Don't get bit by the Cobra Effect

Beyond the Data — Thu, 28 Jan 2021 17:09:43 +0000

This is a short-form video on the Cobra Effect. It is a topic that I think is valuable to understand for those seeking to make better human-centered and data-driven decisions. There are a couple of reasons why this is top of mind right now. First, the Cobra Effect is an important concept that everyone should be aware of, especially if you are in a position where you are creating or influencing metrics and their target. Second, I see more and more gaps where discussion around incentives and unintended consequences is not deliberate prior to actions being taken.

The Cobra Effect can be summarized as incentives designed to solve a problem, but instead, the incentives end up making the problem worse. The name of this came about because, in Delhi, India, there was a problem with venomous cobras while there was still British government rule. The government decided to reward any person bringing in a dead cobra. At first, this sounds like a good incentive to get rid of the cobras.

Unfortunately, a bad outcome came about in Delhi. What do you think it was? Yes, this well-intentioned incentive resulted in cobras being breaded so they could be killed and turned in all to get the reward. Eventually, the government realized what was happening and stopped its incentive. In the end, there ended up being a bigger cobra problem than what existed when the incentive was started. And this is how the Cobra Effect got its name.

All too often, incentives can have unintended, perverse results. Think of incentives given to banks to get people to buy houses to give loans for people to get houses but not really incentivizing that the loans be done in accordance with balancing ability to pay.

I wonder what new and well-intentioned incentives are and will be formulated based on COVID-19 that may lead to unexpected results. If you have any thoughts on this topic, let me know.

There are a number of ways to minimize the Cobra Effect:

Learn From Others: Learn from others providing similar incentives and potential unintended consequences - remember this is highly context-dependent, so don’t assume somewhat similar circumstances should have the same result!
Engage Stakeholders: Engage stakeholders early and get open and honest input. This is not always easy, and if you don’t have engaged stakeholders who can speak openly, then don’t fool yourself into asking those who aren’t to justify actions.
Experiment and Start Small: Experiment by rolling out incentives to a subset for a limited time. Think of experimenting with a couple of locations, a region, a department, etc. Any experiment should be something that is done with good experimental design and created with the goal of true learning, even if the learning is that the incentive was bad and needs to be redesigned.

I hope next time you are looking to design a new metric and target that, you take into account the Cobra Effect and avoid getting bit by it.

Learn more about the Cobra Effect, along with references, here. And please share this post and video with anyone else that you think might benefit. After all, we all benefit from each of us making better human-centered and data-driven decisions.

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

Harness the power of surveys done well

Beyond the Data — Thu, 14 Jan 2021 02:14:00 +0000

What do you feel when you hear these words, “Excuse me, but do you have 5 minutes to take a brief survey on….” If you are like most, then you probably grimace at these words. I’m guessing the last time you heard words like this and had been excited was when you wanted to stick it to them because you were unhappy about something.

Most people hate surveys, and survey participation has been going down for years, but does this mean you should not use surveys? Absolutely not! You should stop doing bad surveys!

Surveys are one of the most important research tools. They can provide meaningful qualitative and quantitative data and do so cost-effectively. You just need to make sure you understand when to use them, how to design and execute them, and how they fit with your other research approaches. This article is going to focus on the best practices in using them.

Survey Design

Start with identifying the ‘Why’ survey: Before creating a survey, you should understand the “why’ you are creating a survey. These are the objectives of your survey, and be explicit. There should at least be some hypothesized value.
What will you do with your survey?: Start with a plan on what you will do with the results of your survey that aligns with your ‘why.’ You don’t have to have every item mapped out, but by thinking through some of this prior, you will have a better opportunity to design the right survey.
Design Your Survey with Your Audience in Mind: Your survey should be designed in a way that you gain insights from your target audience in a way that will help you achieve the survey’s objectives and goals. For example, you may be surveying your customers leveraging Net Promoter Score (NPS) to understand how your software and services benchmark to your peers. Alternatively, you may be a services organization that is looking to expand into new service offerings, and you are surveying your audience to understand what pain points exist now to identify if there are new service areas you may be able to help your clients.
Use the Right Question Type for the Job: There are many types of questions that you may leverage in your surveys. The two general types of questions are quantitative questions and qualitative questions. You generally want to have a mix of quantitative and qualitative questions in your survey.
- Quantitative questions are questions like multiple choice, Yes/No, and rank the following and similar questions. When evaluating the responses here, you will look at things like the mean, median, mode, variance, and other statistical interpretations. The good news is that depending on how the responses come in, there may be a good deal of confidence that can be interpreted without spending a lot of time reviewing.
- Qualitative questions are questions that provide small or long answer textual responses. The best qualitative questions are open questions that encourage respondents to elaborate and share their insights. These open-ended questions are where a lot of the true gold can be gathered from surveys. It takes some time to interpret, but there are tools like sentiment analysis and frequent word usage by question that can be leveraged in some tools, or you can do this analysis yourself.
Throw In Open-Ended Questions for Insights: Open-ended questions can provide some of the best insights you will get from your surveys. Just like good user interviews will help uncover gems through open-ended questions, these similar open-ended questions can lead to insights that you didn’t know to ask.
Keep It as Short as Possible: People are surveyed out. Be an organization that is trusted by keeping your surveys short. The general rule of thumb is that most surveys should be 10 questions or less. There are more complicated surveys that are part of complex research or strong business relationships where there is extrinsic or intrinsic motivation that will offset longer surveys.
Be Honest with the Time it Will Take: There are a lot of surveys that say 5 minutes and, in reality, take 10-15 minutes for most people to take. Understand how long your survey will take and provide the upper range of expectations to respondents sought to be upfront. If they are faster in completing, they will feel better and smarter for doing so and more likely to respond to your survey again.
In order to understand how long a survey will take, you can either have several people take it that are not part of writing the questions or a lot of survey software will provide insight on how long they think it will take to respond.
Use Simple Language: Clear and concise language in your surveys is a must. Don’t make people guess what things mean. This doesn’t mean that you can’t use terms of art for respondents where that term of art is known well.
Use Unbiased Language: Do not use language that will bias your audience. Even using certain words that may relate much more to one segment of your audience than the other can lead to misleading results. Further, if you are leveraging leading questions or examples that prime your audience, then you will have biased responses. If your survey system doesn’t analyze for biased language, then you should have a diverse peer group that can review and help you avoid biased language.
Spell Things Out if in Doubt: In addition to using simple and unbiased language, you want to spell out what is being asked for if there is potential doubt. You might provide an example response or an example scenario that would differentiate what you are looking for and what you are not looking for. The only caution in doing these examples is being careful not to bias your audience by priming them for a response.
Prioritize Your Questions by Building Down, Not Building Up: Start with your most important responses first. You want the most mental attention on the survey spent on the most meaningful questions. Note there is also the potential that people might not complete your survey and leave, so some software will even save those responses even if the user doesn’t finish. Now, if you choose to use these unsubmitted responses, it is another matter.
Only One Question at a Time: Having a compound question means you will have extra response variance and, in turn, less question validity.
Use Smart Branching to Focus the Right Questions with the Right People: Many of the survey providers out there allow for the branching of questions. If a user chooses one response, then the next question can segment the user into another question chain versus another respondent who chooses a different answer. This type of software is especially beneficial in allowing you to submit your survey to a less-known advanced audience.
Segment Your Population Surveyed: If you have several types of customer segments you are surveying, then you can either have a survey designed for each, or you can use smart branching, as mentioned above. The key, though, is really segmenting your audience ahead and understanding what insights you are seeking from each audience type.
Give People an Out: All too often, questions do not provide the user an out. Don’t force an opinion on something if they don’t have one. Having an ‘Other’ or ‘N/A’ option is a wise option.
Ask What People Have Done, Not What They Will Do: When seeking input on what people will do, you understand that by asking questions about what they have done. What have they done to solve a problem? What have they bought in the past? How many times in the last year have they done something? How many minutes do they spend doing a task? All these items relate to items that have been done by the respondent.
Be Consistent with Your Questions: Being consistent in how you present questions to users is valuable. Don’t reverse scales. For example, you don’t have a question 1 (most strongly agree) to 5 (most strongly disagree), and then in the same survey, another question has 1 (most strongly disagree) to 5 (most strongly agree). Don’t change rating scales. For example, you don’t ask one question on a scale of 1 to 5, and then, in the survey, give a scale between 0 and 10 on another question.
Be Consistent Across Surveys: Keep your survey questions consistent over time if you want longitudinal insights. Even slight changes to questions can dramatically increase how people interpret questions and respond. If you change a question and notice an increase or decrease in the next year, then the assumption should be this is likely because of the question wording. Sometimes, changing questions is needed, but you know that if questions are changed, then you are starting your historical data over.

Survey Execution:

How many responses do I need? Whenever determining how many people you need to survey, you need to understand how big of a population you are translating results into, how confident you want to be in your results, and what your expected survey participation rate is.
- Population size: This is simply the total size of the population segment you are seeking to understand. For example, let’s say you have 1,000 medium-sized businesses as customers. You could always sample 100%, and that may be the best strategy in some cases, like an annual customer survey. However, if you are doing new product research, for example, then you don’t always want to be hitting up all your customers each time you are doing product research. Instead, you want to identify the sample size so that you can extrapolate the results back on your whole client base with high confidence.
- Confidence level: This represents how confident you want to be that the differences in the results are not due to random chance. Most people will use 95% for most research cases but may increase this as high as 99% if they want enhanced confidence.
- Survey Participation: We know that survey participation is not going to be 100%. Accordingly, it is good to understand what percentage of people you need to approach or send out your survey to get to your needed survey size. If you don’t know what that percentage might be, then look online at what percentage response rate is typical of surveys like yours. One important note about survey participation size is that low survey participation can often lead to misleading results if viewed from a quantitative lens. However, qualitative learning, even in those low participation surveys, can be beneficial.
There are statistical calculations that are done to determine based on these factors, and good survey population calculators like Survey Monkey and Qualtrics provide. One note is that each of these calculators does not factor in the expected response rate, so I suggest dividing the Sample Size determined by these calculators by the percentage response rate you expect.
Anonymize or Not: If you want the most honest responses, then having surveys anonymized is extremely important. In fact, there is almost no good reason to not anonymize things like consumer surveys and employee surveys. However, when it comes to surveys of business where there is a B2B relationship then there are tradeoffs with anonymizing surveys or not. Having non-anonymized makes sense, and that is in the B2B space where you want to do follow-ups after the survey and/or you want to longitudinally understand how that specific B2B customer views you. Sometimes it is best in B2B relationships to have an extremely brief (oftentimes just Net Promoter Score) anonymized survey. Then, you should have a non-anonymized survey that is more of a longer report-card-like approach where you will have a longitudinal understanding of your customer and be able to follow up.
Get Question Input from Others: Any survey worth sending out is worth getting input from others. Sometimes, this is a colleague with experience as a second opinion, but for larger surveys that are important for organizations, then getting input from a broad group of stakeholders is important. It is easy to discount the outcomes of things you have not participated in developing, so by bringing in stakeholders, you are getting their buy-in also.
Test Your Survey: Always test your survey as a participant prior to sending it out to your survey participants. This includes testing to respond as you think it would work.
Distribution: There are a lot of ways to distribute surveys, from paper surveys in person to emails, text, etc. The key is, again, understanding your survey audience and where they would feel most comfortable getting survey requests. If you regularly communicate with your customers via email, then sending a survey via email with the email providing context makes sense. As part of the distribution, be very upfront on time effort and the ‘why’ behind the survey frame in a way relatable to your targeted respondents.
Follow-Up - Don’t be Annoying: Not everyone wants to respond to your survey. It is best to try between 2-4 times and then stop. Further, if someone asks you to stop after the first time, then you should respect that. As context, B2B surveys where there is not a prior relationship will often be under 5% compared to the typical response rate for surveys, which is roughly 30%, but can vary drastically based on whether B2B to B2C and if there is a prior relationship with the person being surveyed.
Don’t Over Survey: Another way to be annoying is to over-survey your customers. Understand the relationship you have with your customers. If you survey your customer after every customer service call, then that is annoying for most. This doesn’t mean that you don’t provide an option for a person to give feedback if they have it. There are some customer relationships that people have in the B2B or B2C space, though frequent input requests can be valued and strengthen the relationship. In these cases, it is especially important to demonstrate action based on survey responses. This should be done even if anonymized by sending to respondents and non-respondents, so that respondents can understand their insights might have spurred action and non-respondents will be more likely to participate in the future.
Provide Participants a Bonus? Sometimes, it makes sense to use a bonus to thank or entice a potential survey respondent. You must be careful that the primary reason for taking the survey can’t be to receive a bonus, or else your data becomes less meaningful. Further, if you start using a bonus, then you want to consistently use a bonus in similar surveys. One bonus approach that often will help still be enticing but mitigate personal pecuniary bias is donating a certain amount to a charity for each respondent. You may even have several charities highlighted, and the respondent gets to choose which of the charities receives the bonus from the respondent’s participation.

Surveys are a valuable tool that helps you to understand your target audience cost-effectively. Done well surveys will be one of your most valuable research tools. Follow these steps above on designing and executing your survey, and you will be on the right path to a more successful survey.

Need help?

Contact me

Do you believe in product magicians and data unicorns?

Beyond the Data — Thu, 24 Sep 2020 15:57:00 +0000

Two hot or dare I even say sexy roles are product manager and data scientist. Some make the magician or unicorn metaphor for great product managers and data scientists even. No, I don’t believe in real magic and unicorns and certainly not product manager magicians and data scientist unicorns. But, no matter if coming out of undergrad or grad school these roles are highly sought after.

These roles can be fantastic and extremely rewarding but they can also be infuriating and disappointing. Just like any role you need to line up a person's ambitions, a person's skills, and market's demand and where they align is probably the best fit.

However, product managers and data scientists often express frustration. This frustration is sometimes because of the misguided expectations of the person doing the job, and sometimes because of the misguided expectations of leaders on the purpose and expected outcomes.

This write-up provides a little expectation guidance for leaders seeking out product managers and data scientists. And, it provides aspiring product managers and data scientists guidance on what to expect.

Product Manager the Great

What is a product manager if not a magician? A product manager is responsible for the strategy through execution of an organization's products (digital, physical, and/or service). The product manager does this by identifying customer and market needs and aligns these with organization strategy. Then, the product manager coordinates various teams and resources to develop, launch, and enhance products that meet these customer, market, and organization needs.

Not so hard this may sound like. You couldn’t be farther from the truth. Product manager is a punishing role because of the complexity and demands the role faces day in and day out. Add to this that most product managers have influence but not authority with the teams they need to coordinate but are often held directly accountable for misses of others.

In my experience, the people that are the best product managers are people that are:

Curious: Curious people want to understand and learn more. They uncover what others don’t. Ask questions that others miss. In my experience this is a trait that people must have coming into the role and can’t be trained.
Empathetic: Empathetic people understand and feel others. They are good listeners. They want to understand others. This doesn’t mean that they are pushovers and must make tough decisions but they can do so with understanding. In my experience this is a trait that people can certainly develop but it doesn’t hurt to have natural tendencies towards empathy.
Trusted: Trusted people naturally communicate with others in a way that gains respect from others. People want to help them and will go beyond the minimum for trusted people. They will also waste less time bickering on things that are irrelevant out of a lack of trust. In my experience this is a trait that people can develop. Of these traits trusted person is the trait that takes time to build in the eyes of others but also can be quickly lost.
Communicator: Clear communicators can provide updates, insights, and in-depth analysis to others in a variety of forms. They leverage their empathy and trust and provide relevant messages to their audience. Clear communicating product managers are not rare but often product managers are not good across multiple forms and have a challenge to provide the executive summary.
Team Player: Product management is a team sport and accordingly being a team player is essential. This means understanding your role and how it plays within the team. Working well and supporting other team members. Stepping up when it is your turn to step up and playing a supporting role as needed. Ok product managers can step up when it is their turn, but great product managers also know when to play a supporting role. This includes checking the ego at the door which some product managers have a challenge doing, but not truly great product managers.

There are certainly other traits and skills you want your product managers to have but these five standout for me.

As a product manager these traits become really important when a lot of your time is spent (or should be spent) understanding and communicating with future and existing customers. Understanding customer need and aligning to your company’s strategy is where a significant amount of time should be spent. There is also significant amounts of time building relationships and engaging with internal teams like engineering, design, information technology, marketing, sales, etc. as a product manager.

The Mystical Data Scientist

Are you telling me the data scientist unicorns don't exist? A data scientist is responsible for answering difficult organization questions that require the combination of technical and analytical skills tied with a deep understanding of the domain. A data scientist often coordinates with others ensuring adequate data and other resources exist to answer these complex questions. Further, a data scientist performs this complex question analysis, often as part of a team, and communicates the results of the complex question analysis to others.

In my experience, people that are the best data scientists are people that are:

Curious: Curious people want to understand and learn more. They uncover what others don’t. Ask questions that others miss. Analyze things others don’t. In my experience this is a trait that people must have coming into the role and can’t be trained.
Empathetic: Empathetic people understand and feel others. They are good listeners. They want to understand others. They deeply understand problems. In my experience this is a trait that most data scientists are lacking despite how much value it can add.
Trusted: Trusted people naturally communicate with others in a way that gains respect from others. People want to help them and will go beyond the minimum for trusted people. They will also waste less time bickering on things that are irrelevant out of a lack of trust. In my experience this is a trait that people can develop. Of these traits trusted person is the trait that takes time to build in the eyes of others but also can be quickly lost.
Communicator: Clear communicators can provide updates, insights, and in-depth analysis to others in a variety of forms. This includes skills around visualizing data in a meaningful way. They leverage their empathy and trust and provide relevant messages to their audience. Clear communicating data scientists are well respected because they are anomalous unfortunately.
Team Player: Data science is a team sport and accordingly being a team player is essential. This means understanding your role and how it plays within the team. Working well and supporting other team members. Stepping up when it is your turn to step up and playing a supporting role as needed. Being a team player means not only within the analytics team itself but also the broader team that is tackling business problems and data scientists will have a number of teams they play on at any time.

There are many other traits and skills you want your data scientists to have including good technical skills but these five standout for me.

As data scientists these traits are important so you do the right analysis and get maximum impact from it. Reality is data scientists spend more time data cleaning than anything else. Even organizations with good data engineering practices still have data scientists doing a lot of data wrangling. Certainly great data scientists spend more time up front making sure they are not working on the wrong problems or misunderstanding the problem. Great data scientists also spend a significant amount of time developing and communicating out results in an audience-centric manner.

Lots in Common!

Ok, this might not provide the right message. ;) But, it was too perfect from when I was in Lima, Peru!

The interesting part of course is the key traits that great product managers have are the same that great data scientists have in my experience. There are huge differences in what these two positions are responsible doing and they each would have an extremely hard time switching roles. But, the similarity in key traits is maybe not so surprising because in general great product managers and great data scientists are people that can identify problems, apply information to make a decision, and clearly communicate out findings in a trusted manner.

For both product manager and data scientist you don't notice things like CEO of the product or data, data unicorn, or product magician. But, both of these roles clearly require complex skills and broad understanding of many disciplines. Further, they both require the ability to align and leverage resources across multiple teams that result in making important decisions.

The challenge is since the titles product manager and data scientist have garnered attention by both people wanting to be in the role and leaders that want their organization to be successful there is a belief that these positions alone can deliver value. These roles are merely parts of an ecosystem. And, if delivering successful products then you need successful customer service, engineering, design, marketing, sales, and many more teams working together. Yes, the product manager has an important role in delivering successful products but organizations are only as good as their weakest link. The same is true about data scientist and being able to help make complex data-driven decisions.

Accordingly, if you are a leader setting expectations related to these roles don't look at product managers or data scientists as people to "paper over" dysfunctional organizations and garner success. And, if you are or seeking to be a product manager or data scientist then you should have the perspective that you are part of a large ecosystem and the success of a product or data-driven decision is a result of the team's effort. Being a good product manager or data scientist in a healthy organization ecosystem will often deliver amazing products and make great data-driven decisions.

Hopefully this article has made you laugh a little, think a little, and maybe just maybe alter your expectations for the better around product managers and data scientists. And, if you happen to be a product manager or data scientist reading this then maybe you have a new-found appreciation of your colleagues with the data scientist or product manager title. As always, feel free to reach out and let me know your thoughts.

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

Ep 43 - Michel Guillet - Leveraging product and data-driven thinking to enhance sales teams

Beyond the Data — Wed, 23 Sep 2020 13:30:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“I think that one of the main keys is to really focus on the workflow. Workflow first, like what are they trying to accomplish? What’s their day? What are they doing every day, week, etc.?”

— Michel Guillet

Michel Guillet, Sr. Product Manager at SalesLoft

This was an episode where I talked with Michel Guillet a Sr. Product Manager at the fast-growing SalesLoft startup where they help companies engage with customers, build pipeline, and close revenue, faster and we discuss how human-centric and data-driven approaches can help sales better succeed.

Sales people and sales managers are constantly under pressure to close deals and reduce executive revenue uncertainty despite an unpredictable world and the pressures of life and work. Michel and I talk about some of the challenges sales people and managers face and how human-centered and data-driven approach can help generally including some of the things SalesLoft is doing to help its customers.

More about Michel:

SalesLoft: SalesLoft
Michel’s LinkedIn: in/miguillet

Transcript Download

go Beyond the Data podcast

Ep 43 – Leveraging product and data-driven thinking to enhance sales

Michel Guillet of SalesLoft

Machine Generated Transcript via Descript

Dave Mathias: [00:00:00] Hey everyone. A couple of quick items before getting started today. Really good to have you here. Of course. today we're going to be talking with Micel Guillet. It's gonna be great interview so I hope you get a lot from that, but before getting into the interview, just wanted to mention that you will now find this as go Beyond the Data in your feed. That isn't a mistake.

Go Beyond the Data is all about using data to make better decisions, but really going beyond the data in leveraging things like behavioral science, product management, customer experience, user experience, and other things that organizations are trying to leverage to change the status quo, make better products, have better insights and have better experiences. So we're not gonna just be covering things around data.

Now let's get into today's episode. Today I have Michel Guillet with me and Michel is an awesome person that has a ton of experience. He has 25 plus years experience leading the design and implementation of data and reporting solutions.

Hey Michel. Good to have you here.

Michel Guillet: [00:00:56] Hey Dave. Good to spend time with you.

Dave Mathias: [00:00:58] And your experiences, includes, working with structured and unstructured data across digital advertising, consumer goods, healthcare, insurance, like tons of industries. You're also a senior product manager, your day job. You're senior product manager of analytics and data science at this, this little fast growing startup.

That's not so small now, called SalesLoft in Atlanta. We'll talk a little bit more about that, but you're also an adjunct instructor of data viz and presentation, which I love the presentation part, not just the visualization at Georgia State University - Robinson, College of Business. We'll get into that a little bit more too cause I think that you'll have a good perspective also from, being in the academic world, in addition to being in the professional world. Good to have you here with us today.

Michel Guillet: [00:01:37] Yeah. Thank you, Dave. I'm excited to spend time with you. I enjoyed our conversation a couple months ago, so this is great.

Dave Mathias: [00:01:42] This is awesome. and so we're talking on a Friday morning and love to learn a little bit more about you. I know more about you. You're really interesting. Hence why we're having you on here, but tell us a little bit more about how you got to the career point where you're at right now.

Michel Guillet: [00:01:56] Sure. I'll just give you some snippets. My undergraduate was really work was in international economics where I thought I would work on a crop and agricultural issues in West Africa. My obvious French name gave me some interest in kind of, leveraging West Africa and the French nation and doing that.

And I, my first job out of college ended up being with Estee Lauder, where I got to help with a lot of the pre EDU transition, doing a lot of inventory analysis, hands on the fact. I dated me a little bit, but I started using, the Excel predecessor Lotus one, two, three in 1988.

Dave Mathias: [00:02:30] Oh, wow. Ouch I do have some distant memories of that. How does actually, as a kid, it's not good memory, so I can tell you.

Michel Guillet: [00:02:39] Know. Through no, a little bit of academic background, but just through my own training, just became a little bit of an Excel. Was it real be pan and throw through graduate school Anderson, Andersen consulting work, et cetera, just continually refine that. And, over time started to lead teams and, spent a bunch of time, doing audio analytics.

Hence the unstructured Mike, I spent time with speech works, nuance and speech recognition. we looked at largely audio analytics in a big way. And then right after it was at nuance that I largely made the jump to product management. And in a, my niche in the world is not really product management.

It's really kind of data solutions where organizations need assistance, making those data solutions valuable for their customers. If that, as that is my niche in the world, or at least for the past 10 years,

Dave Mathias: [00:03:29] Yeah. that niche is getting to be a bigger, important part and less niche-y as we've gone to, obviously digital products have become a lot bigger. And at Sales Loft tell a little bit more about Sales Loft and what you do there, what the organization does. It gives perspective to you and what you're doing now as a senior product manager in analytics and reporting.

Michel Guillet: [00:03:46] Yeah, Sales Loft is great. So we are, cloud-based, sales enablement platform. If you think about contextually, like you have marketing automation have Salesforce and CRM, we are basically the system of execution. We are the way that sales folks do prospecting deal engagement, and then customer engagement.

so it's relatively new and the space, but we're an eight year old company growing rapidly, serving the needs of Small companies that need to sense a sell as well as large enterprise organizations. the part that gets to me excited about being at Sales Loft is our mission is really focusing on changing the selling and buying experience for both the seller and the buyer.

Not just producing sales software, but really focusing on experience, in our kind of our team's role, really, I get excited about our team. Our team's role is to really facilitate kind of the display and presentation of information to salespeople and sales managers, so that they know what to do and how to better, run their organizations and be more efficient.

Dave Mathias: [00:04:46] We'll get more into that. we're going to be talking a little bit about design data solutions, especially for those non-analytical customers sales, both on the buyers and the sellers are obviously fall into that category. What is the thing that you find most interesting about your job?

What gets you up and energized every day for your job?

Michel Guillet: [00:05:02] You don't I think what gets me has always gotten me excited over the past couple of years where I was doing healthcare work with nurses is, getting people who don't consider themselves analysts excited about data, right? When they have those Eureka moments of those small Epiphanes like that, they get something and they know what to do that's clearly what good gets me excited.

Dave Mathias: [00:05:21] We might as well get some tips out here and get your wisdom right now. So what is the biggest tip? if you're a person that's got a similar role or similar challenges yourself, what would be your biggest point of advice to get people excited that are non data people?

Michel Guillet: [00:05:35] So I think that as data people, often we think about the data itself. And I think that one of the main keys is to really focus on the workflow. Workflow first, like what are they trying to accomplish? What's their day? What are they doing every day, week, et cetera. As an example would be in sales, like there might be a weekly sales meeting, and just subtle things like, how do we make that meeting successful?

Dave Mathias: [00:05:56] Yes. And not so painful for the salespeople and productive.

Michel Guillet: [00:05:59] Yeah both at Sales Loft and previously at applied systems in the insurance industry, this idea about like, how do you facilitate the right conversation between the sales manager and the reps, or the sales development reps, to be able to have a fruitful, productive conversation so that both parties, walk away saying, I know what to do. I know what to resolve. I know what the next steps are.

Dave Mathias: [00:06:19] Yeah, that's great. And so let's dive into this a little bit more when you're talking about designing data solutions for non analytical customers, how do you start with that approach? When you think about things in the sales, let's talk in the sales segment cause that's where you're at now. And I've had a fair amount of experience in that sales segment too. How do you go about designing data solutions for those non-analytical customers? Like a salesperson.

Michel Guillet: [00:06:39] so obviously one of the reasons is charting or summarizing. The data becomes, a point is that the trade off is understanding, like what is the best chart to display this information versus what the chart that's familiar that may be less productive, but they're more likely to use. Alright, so a four quadrant or a scatterplot or a bubble plot may meet with multiple axes.

Three metrics might be the ideal way to display some information. However, a horizontal bar type effectively might be the way that they could look at it and not have all the information, but enough to be able to. so I describe it, Dave, is this idea about trading off between the productive versus the familiar and striking a balance between those two things?

Dave Mathias: [00:07:20] So I think one of the things that we like to show is we like to show too much data to people, especially non analytical folks and showing people the right information at the right time that is meaningful action, but balancing that also where you don't want to make people robots, you don't, we don't want our frontline people, no matter what the positions are, sales, certainly that they're just robots and certainly having a process of understanding the process and executing on it. But we also, being able to really understand sort of first principles thinking and being thinking, understanding why things are happening and being able to make adjustments on the fly.

How do you balance that with. getting information to them at certain times. I think that's always the challenge, cause I think we're facing that in a lot of our lives where it's like, how much is it are just, we're dictated by our calendars, for example. And our AI is saying, Hey, this is going to be the most efficient people to follow up with and I've just booked your full day as a sales person, versus how much you have the salesperson think or, they feel like they're just picking up the phone.

Michel Guillet: [00:08:16] Yeah, great point. Let me break that down just a little bit. the idea that we're first aligning to workflow means that we're not adding to their work day. We're basically trying to make what they're do already better. So that's the baseline start. The second is that understanding kind of context about what information is right in that context, right?

So you could be saying like, this review is a QBR, but which could be a quarterly business review. So obviously can I compare this quarter to the prior quarter? That's a, that would be table stakes in that context. So I think context becomes a second big piece. And then the other part, I would say, David's the delivery, right?

Is it purely audio? Is it actually a presentation on a TV? what exactly is the delivery of the media that's being used to convey that information? I think lastly is to where you were going is determining like the degree of a prescriptive solution versus exploratory, And the idea that you can be very prescriptive in saying here's the cool, the default, but then can I then change? Can I use interactivity as a way to facilitate exploration? And then second secondarily is that can I have access to it details in case I want to be able to go and do some ad hoc analysis.

Thinking about those, I think the part where people go wrong is that they either go one extreme and saying, what I hear is, I definitely have heard this before in consulting and elsewhere is that. I talk to 10 customers each of the 10 send 10 different things. Let's just give them the data, right?

No prescriptive, no guidance whatsoever. Versus having the, kind of the ability to glean to say, what is common and across these 10, that they're all asking for. And what can we, what can use that as a starting point?

Dave Mathias: [00:09:53] Tied with this I, that I think a little bit, you can push back if I'm maybe jumping a little bit farther down, but, the tendency. behavioral design. so when I think about when you're, especially for sales, for example, is, designing software that's going to try to get to the optimal efficiency for salespeople in general, versus how much do I understand more or I try to understand the tendencies, the strains of that person. And I'm trying to get the best out of that person. Because we're not all the same. We all have different tendencies, we all different likes different things. So how much of it is, software adapting somewhat to the individual and providing insights versus trying to just pull people to a certain baseline or certain sales person type persona types, action type.

Michel Guillet: [00:10:41] Yeah. so this is a great tee up for data science. And I think what you described now is where we think about data science as being what, here's the static report, the prescriptive guidance guided narrative that we're providing for both David and Michel, but what's different about Michel and what can we enhance through suggestions in some way to differentiate because there's something contextually. It may be not the right metaphor, but it's one, we think about a little bit, is this idea about the Netflix recommendation. We all use the same medium but being able to tell me that, by the way, other people, like you also did this. Giving that little slight nudge or that suggestion to a salesperson to be able to guide them beyond what kind of the prescriptive solution is. I think that's where that sweet spot that we want to go to.

Dave Mathias: [00:11:24] Yeah, that makes a lot of sense. So tell me more about what you've seen as successful in this space and developing software, not just obviously yet SalesLoft right now, but also consulting in other places and from my, from a perspective of being a senior product manager in that space.

So you as a senior product manager, you're really focused on aligning, the business requirements where the business strategy to what the customer need is and then obviously, yeah, leveraging data and analytics. Tell a little bit more how that intersection works in what you're doing.

Michel Guillet: [00:11:58] Sure. I think, I'm biased, but I think my academic training as an economist kind of lends itself to being able to Thinking about things from a marginal utility perspective, what is the, where do I get the biggest bang for the least amount of effort? Very traditional kind of product management techniques.

But when you think about it in the context of delivering yeah. Information, it's even more important. I don't know if you're familiar with Amanda Cox she leads the team at the New York times. She is one of my favorite human beings. both in terms of her intellect, but also her, sense of humor. And she's has a quote from years ago that says, data is not like kids, it's okay to have favorites. All right. And what I take away from that is always like what makes sense, which what parts of the data are going to get my users to take action. And they get them to sit up and take notice and want to do and be motivated to do something there's contextual pieces, which are usually secondary.

What are the three to five pieces of information. And I think that ends up being really the starting point. What are the three to five pieces of information that we would change in salesperson's life? Am I having a good day or a good week or a bad day or bad week? What are those to five that are most critical to me? And then everything else becomes a contextual assist. How we deliver that after that?

Dave Mathias: [00:13:09] One of the things you were talking about earlier was also from a leadership side sales management side versus the frontline. how much does your software, thinking about those very different personas, but the objective should be aligned or hopefully is aligned the goals of each. How does your software treat those different audiences different? Or do you think these audiences should be different? How do you go about that?

Michel Guillet: [00:13:32] Yeah, that's a great question. I think that we think about these distinct roles. I think we need to do a much better job on it, frankly. I think we're on that really good path to just saying what is the information that a manager needs every day versus kind of a, see our sales development rep, what, or how do they, should they start their day?

So I think largely that starts in the UX research side. To better understand what those things are. And that becomes a premise about how we do that. I think how we deliver it, that is making the software in a very component way. So if we were to think about in terms of dashboards, the panels on a dashboard might be different for every user will based on their role.

We might give that we might actually be prescriptive in terms of, Hey, here's what we think you should see at the top. Here's what should the detail of the bottom, but potentially also give them the ability to configure it themselves.

Dave Mathias: [00:14:19] Yes. Yeah. that ability to have some choice in some adjustment. And so what things have you seen most challenging in that software space, for sales, for adoption purposes, where they're not resonating with the data. It's not working at the level of what are those challenges that come about. And how do you think they can be overcome better?

Michel Guillet: [00:14:38] yeah. great question. no different than anywhere else, right? How do you highlight for our customers or users when clearly the problem is their data, right? Their data is a little bit off, right? Is there a way for us to gently kind of surface to be like, Hey, the data is incomplete.

And you're not going to get the full picture in that way. So I think that's the kind of, because that will be the difference among our customers. Some have very raw, they have a lot of resources, a sales, operations team, strong administrators, where the data quality is really strong. There they're constantly improving the quality and those that are strapped where they have multiple people wearing multiple hats.

How do we help those users, how do we help those customers so that they, because they don't have some of the administrative resources or capability, how do we account for that a little bit, something we're still wrestling with the honest day, but I think that's the right way. I know that in a prior life, in the insurance space, we basically knew that 13 ways data could go bad.

Right missing data misspellings, et cetera. And we would basically have a little bit of data science, mostly regression type things to make, to run a report for our team internally to suggest, Hey, here are the things that you should recommend to our customers in terms of what's where are there opportunities to correct their data might be.

Dave Mathias: [00:15:52] So related to that is the communicating uncertainty. And so part of that uncertainty is going to be obviously higher if your data's not as good. And so how do you communicate that on certainty and where you might even communicate to help the motivation. To clean data more might I also lead to that.

Where, why are we so uncertain about this, two quarters from now and things like that.

Michel Guillet: [00:16:13] Yeah. So there's some, there's a little bit of a day of science, so there's a little bit of applying some scoring methodologies to be able to Hey, I'm looking at various set of deals. You know of all these deals giving us general nudge your suggestion to me like, Hey, this information is on this deal.

It's a little bit incomplete compared to or others. Alright. So just as an example, the second one would be is just really I think one of the things I love about Sales Loft, we've made such a serious investment in design and UX and research. And I think that not only having designed concepts, the testing, those things, to see that we're getting the right feedback from customers and users about Hey, do you understand what this is?

Oftentimes it on the date of this side, we use color as a way to do that. But. Design, our design team has many other techniques to be able to convey, not just uncertainty, but just Hey, incompleteness.

Dave Mathias: [00:16:58] That makes a lot of sense. You've obviously covered a lot of industries and you, as part of that, one of the things that I know that we were talking about before we actually started pressing record is just the importance of curiosity for folks that are in either in data or in product or space like that.

Talk a little bit more, we're even talking about different ways to test for that. And, just thinking about, from our different experiences, talk a little bit about how you viewed just a curiosity in these spaces. How do you, what do you look for that? how do you encourage that?

certainly there's traits that people are more curious than others, but yours also, I think there's environments that encourage curiosity too. Cause there's curious people that get stuck in environments that really don't encourage that. So it gets stamped out. Some is part of the problem.

So can you talk a little bit more about what your experience is and how to both create an environment that has good curiosity and also how do you, help discern. Is this person going to be a good person? Cause they're going to have a good level of curiosity as one of the items.

Michel Guillet: [00:17:56] Yeah, let me talk about it two ways, if that's okay. Dave, you want to talk about in the context of a team and an analyst team and a development team, and then also in the context of users, in the context of a team, like if we had an analyst team or I've had analyst teams in the past, curiosity becomes really tied, strongly to.

Your ability to clarify, not just the problem, but the big problem. So as you can imagine in a sales context, that might be like, what can we do to help our customers grow their revenue this quarter? You need, you require curiosity on the team. Even if the team doesn't really know. a lot of our dev team.

Has never done sales, but how do we facilitate a curiosity in them to under better understand sales? And I think it really goes to framing a bigger problem, not just a tactical, Hey, we need to put a number on this screen. So I think that's one way and I think it's also related to the team as is.

Giving them the opportunity to like, get excited about that. Hey, you're gonna have, one of the things we talk about with teams, so it's like, Hey, this is going to impact greater than 30,000 users next week. And I love the fact that our team gets excited about doing that.

They don't shy away. They don't shy away from it and shy away from a big problem. They don't shy away from having a big impact. So I think that's curiosity can be largely rooted professionally in that way. I think that when it comes to users, it's really this ability to you think of the visual cue I would give everyone is just like this idea of an ellipse, right?

There's a little bit more right. Give them a little bit of a nugget, but give them a place. If they want it, they can go more and then take an interest from a product analytics standpoint. Like where are they asking for more? What part of the app are they asking for more? Are they more curious about that?

But I think there's always a path to being able to give them a, just a little bit more. in fact, in some ways I think it's, we make the mistake of putting too much information on the page very often versus saying, let's just give a few pieces of nuggets of data, and then if they need more, there's a place to get to it.

Dave Mathias: [00:19:47] Yeah. And I think that more, but also in prioritizing sort of the impact, the, again, that big goal that your, the user's having to. and so I think what I've seen a lot, especially in the sales context is on the sales management side. There they'll slice and dice, as soon as they get. Something to start looking at, they'll just find, and they'll find what they want to find almost, where you think it's going to be a tough quarter.

You're going to find as much, many reasons why it's going to be tough. And so how do you help users get out of those? Some of those natural cognitive biases that we face that are probably not either overly confident or overly negative and both. And I think a lot of these, just like being an ass fleet, I think salespeople, I always think of A good salesperson oftentimes feels like that because there's so many ups and downs and you're going to have the wins, but you're going to have so many losses and part of it's grinding it out and showing up and trying to keep that good spirit throughout.

So how do you, help that from a datas, from a sales product side that can help those salespeople, whether it, but also helps the sales managers, where they have obviously their own pressures on that front.

Michel Guillet: [00:20:50] Yeah. I think that the nature talking to customers and even internally on our services team is convinced them that we don't have, we may not have the answer, but we're low. Our goal is to give them the ability to ask better questions, right? To drive them to a point to be like, I don't have to, I have to go give Dave some bad news because his performance is underperforming. Give me the right information. So I can go ask Dave the right questions about what's good. The other, this is something I've I there's little bit of debate in kind of the community about whether data Abbott tells you why skewed towards the idea that data really never tells you why.

Okay. So you won when, how, what, et cetera, a conversation is the way the Y happens, right? That's where like Dave had a crummy week. I don't know why they've had a crummy week, but I can at least better understand, like Tuesday was awful. let's go talk specifically about Tuesday, Dave, tell me about Tuesday.

Dave Mathias: [00:21:41] Yeah. it is interesting cause I do wonder how much in the AI front, how much we want to more encourage that behavior in those conversations and trying to not just over be overconfident in that we see these numbers that this is, not that though. We want to ignore the data, but we want to use the data in addition to those human conversations and not.

And my fear is that we're moving too quickly and we want to go down that path too quickly. So

Michel Guillet: [00:22:07] Yeah, so sales to me. And the way we think about sales and Sales Loft is all about the relationship, right? It's about improving that experience. So the idea that we would decouple, Have information decouple that is, would be different, right? The idea that we should support those conversations.

Dave Mathias: [00:22:22] Definitely. Definitely. I know we're running up on our time a little bit, but one of the topics wanted to hit on a little bit more is data ethics and product managers. And it's certainly a big topic nowadays and love to. from a, from both a product person and a data person like yourself, and we're both fall into that category.

What is your view of what responsibility product managers have around data ethics? What can they do to meet those responsibilities and such.

Michel Guillet: [00:22:48] a little plug. I have a workshop at, GSU for data ethics, where I work with data scientists, getting them to the way we frame it up or the way I frame it up is, it's not, if you're going to be an ethical conversation about data it's way. Okay. So broadly what we do with the students and I do with our internal teams is we're developing the muscle.

We're practicing and thinking through scenarios to develop that muscle so that when that real scenario happens, can we're ready? I think broadly for product managers. Obviously our legal groups are going to keep us honest from a privacy standpoint. I think the two places where things go awry or one on the fancy word called Providence, which is where did the data come from?

we scraped, we were in you're in a startup. We scraped it off some of the site. You know were we allowed to do that? Any risk there? Secondly, so Providence becomes a kind of a concern from the product manager side. Secondly, I think particularly in the context of AI and data science, that the biggest risk that all of us face is, the issue that we're going to negatively impact one audience. That audience could be small. It could be. So for instance, if you view sales off an example, Oh, this only affects 1% of our users. 1% of our users, it could be 300 people. I think that we have to be aware of those smaller communities or the smaller groups of users.

and if you're in if you read all the stories about where AI go goes bad from an ethical standpoint, it's not as impacted the majority. The bank loan who did get a bank loan it's who didn't get a bank loan. So I think that's the big thing that we have to be really focused in on is understanding our product manager training is very much the marginal unit.

most people will get the benefit from this. But particularly when it comes to ethics and data and AI, I think that's the risk. We have to be hypersensitive to.

Dave Mathias: [00:24:33] And so is that risk to be born by the product manager or how much is it just the product manager engaging legal? cause I do wonder on some of these things where I think we're trying to make product managers to too much for unicorn type of role and how much can they really understand things like the CCPA and the GDPR and all these other things.

And they should understand honestly, basics of those types of things. but, what's fair use for certain scraping and like what way you're using. one thing that I've always recommended for product folks is just to engage the experts early in whenever. And just your biggest role is in that standpoint is issue spotting.

Like you need to be able to spot. there could be an issue here and getting people that are really a lot more versed in that than you. certainly if you're in a small company, you may have to wear so many hats and probably not do a great, but, at the same time, I think for, especially when you're a more sizable company, it's just a matter of getting people involved early, in that, is that fair to say? Or what are your thoughts?

Michel Guillet: [00:25:31] Yeah, I think we're thinking the same way. I think that product manager has to own the conversation, This conversation needs to happen with legal. I need to own the conversation and make sure that the conversation happened. We document it and the decision we make collectively as an organization.

But I think the product manager needs to own the conversation and make sure that it happens and not avoid it.

Dave Mathias: [00:25:49] Yeah. The key is a lot of these conversations early, before a lot of investments spent in that it's wait a sec, we can't do this product, that we were depending on that we spent all this money developing, seeing that, too often. Okay. any other words of advice. For a person that's a, in your case, more of a data person turned product person.

If somebody would want it to follow along in a career path, similar to yourselves, what advice would you have?

Michel Guillet: [00:26:10] Feel comfortable knowing the customer, I think for a lot of technical folks or data folks, the idea about, bias confirmation bias becomes a big issue. Oh, I know what you need because I know what this is what I would use. So I think on a product, the data person going into product person, the biggest risk is the confirmation bias.

You like, I know exactly what you need. And I think you have to walk into it saying. All right. I'm a complete, like for me right now, I've had experience working with salespeople previous lives. I have to go into the conversation. Like I don't understand sales, but explain it to me as if I don't know anything.

And so I think that's the best advice I think, on the reverse product manager, wanting to know more about the data is I think just thinking in the context of a basic Excel, right? Those Excel metaphors play out really well. Yeah. if you can mock it up in Excel, if you knew how to create an example in Excel, we're just talking about scalability and sophistication, but if you can talk the talk in Excel, you're more than half, more than halfway ready.

Dave Mathias: [00:27:06] I know that those are both great piece of advice. We're going to hit a few. actually, one of the questions I want to ask, we're not going to end on this, but were you, is what is your biggest failure that you think you've had in your career that shaped you where you're at today and what did you learn from it and how did it shape you in a positive way?

Michel Guillet: [00:27:23] So I can I'm sad to say. I've probably made the same mistake a couple of times, and I made it a few times to get it right. There are a couple of times others raise concerns to me about data, and I knew the data and I knew the situation and I made him, I said, I'll give you a little detail. So there was a piece of some information for a large bank that we were working with.

The information was wrong, but I knew the information would be wrong, but I knew it wouldn't be significant. So if you think about a large bank, we're talking like, it was call center data. It was tens of thousands of users, right? The data was off by 4%. I knew it was off by a small number.

It's an insignificant, they weren't going to change the business, but the customer and ironically the service people, they were very concerned. And my body language and my attitude did not reflect the urgency that they had. So I knew what to do. We fixed it within a day, but I think in multiple cases that like a turret, my, a little bit of my credibility, sometimes, I've been in, I tend to not, I tend to not panic or get worked up about things.

And that sometimes that works against me.

Dave Mathias: [00:28:23] Yeah, that's great advice. so certainly you faced that a few times, but we're talking about how to have greater empathy for the audience. you're talking with them. What ways have you been able to learn and apply how have you been able to change some I'm? sure you've obviously recognized it. So first step is getting over denial. So admitting, can you, tell a little bit how you've adapted, based on that.

Michel Guillet: [00:28:46] Yeah I think my interviewing techniques, I'm playing a lot more advanced interviewing techniques that you would use in product management, but really restating the problem back so that I have their perspective and it's not mine. So that becomes the easiest way. Oh, The meeting, you're going to use this information in the meetings wrong. That meeting is really important for these reasons that I get that right.

Dave Mathias: [00:29:06] That's great advice and something. Everyone should be taking the count if you're not doing it already. Yeah.

Michel Guillet: [00:29:10] And data nerds tend to have a little bit of a purse, especially when you're talking with non analyst, you can come across as arrogant or others are concerned about, Not saying the right thing. So there's already a level of anxiety when the non data person's talking to a data person in some cases, and you have to you have to be able to work with them versus take your own perspective.

Dave Mathias: [00:29:29] Yeah. using things like not using acronyms and trying to create an environment, that's more equal playing field that you're not, trying to, show things in let's go draft this dashboard together using. Tableau right out of interest, do it together versus let's just go to a whiteboard and those types of things.

I think, ways that you can try to create greater parody. So can I ask you a couple of quick rapid fires and then we'll go into wrap up. Cause I know we both have things coming up, got to ask, since you are a data viz instructor, as one of your things as a, at the university is if you were a database, what date of his would you be?

Michel Guillet: [00:30:01] I would be a, sparkline a simple, single purpose, very focused, one goal only that would represent me. My wife would probably say a scatter.

Dave Mathias: [00:30:13] Is he, at least my wife would say scatterplot. And honestly I say I'm scatterplot too. Cause I've admitted by now.

Michel Guillet: [00:30:20] maybe I'm still in denial.

Dave Mathias: [00:30:22] what is your favorite book? it doesn't have to be a data or product or anything, but it could be just pure pleasure book. What's your favorite book?

Michel Guillet: [00:30:28] Yeah. it's, it changes, years ago, my mom gave me the book Small Is Beautiful Economics as if People Mattered. So that was really her way of getting me to think things. So I think that had a really strong influence on me early on. I think there's two things right now.

The Checklist Manifesto by Atul Gawande, is my favorite business book and the one I recommend to folks the most and then pleasure reading goes, the Drunkard's Walk. It's how randomness rules our lives is something I recommend as well.

Dave Mathias: [00:30:58] Great that's a few recommendations. And so before we go how are people gonna get ahold of you? Where should they follow you? That kind of thing. What's the best spot for them to, if they want to learn more, want to connect with you?

Michel Guillet: [00:31:10] I think LinkedIn is probably the best way.

Dave Mathias: [00:31:13] So on linkedin that's M I G U I L L E T so M I G U I L L E T looks like the right way to get ahold of you then .

Michel Guillet: [00:31:24] I am on Twitter, but I generally just use it to follow folks that I'm intrigued about. or when I'm looking to learn about, for instance, I'm deep in some machine learning Things I'm following a few folks on the machine. So I'm using it as a way to consume information, not necessarily to share or publish, but LinkedIn, I am more inclined to share or interact with folks.

Dave Mathias: [00:31:42] Excellent. So we'll put the link out there for your LinkedIn and hopefully some folks will connect with you So any other, things you wanted to hit on before we drop off?

Michel Guillet: [00:31:51] No, this has been wonderful. Thanks Dave, for kind of making this happen. I really appreciate it.

Dave Mathias: [00:31:54] Yeah, great talking with you again and hope all is well, you have a great weekend ahead and we'll talk to you soon.

Michel Guillet: [00:31:59] Thank you, Dave stay well.

Learn more

Permalink

Ep 42 - Jansen Sullivan and Victor Anjos - Data education deserves better

Beyond the Data — Thu, 10 Sep 2020 14:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“I you’re largely doing really strong reporting and visualizations that tell a story and deliver business value. You’re going to win and have a really good career. All throughout. Those are positions that are largely needed and hugely available as opposed to the data scientists that everybody believes.”

— Victor Anjos

Victor Anjos

Jansen Sullivan

This was an episode where Jansen Sullivan and Victor Anjos of the All Things Data podcast by 1000ml and I talked about data education and some of the gaps that we hope get resolved. A lot of progress is being made but lots more needs to happen.

Jansen, Victor, and myself all believe that traditional analytics education moves too slowly and is often too focused on checking boxes more than creating good data analysts, data scientists, data engineers, data product owners, etc. There is a lot of great efforts that traditional universities are working to adapt but the system lacks the agility.

Beyond the traditional education approach, the analytics space is a space that requires real projects and experience with end-to-end problems to advance people’s skills. More apprenticeship learning is required and love to see more organizations recognize this and support this.

More about Jansen, Victor and All Things Data

All Things Data Podcast: https://1000ml.io/category/podcasts/
1000ml Website: https://1000ml.io/
Jansen’s LinkedIn: in/jansensullivan
Victor’s LinkedIn: in/victoranjos

Transcript Download

Data Able Ep 42 – Data education deserves better

Jansen Sullivan and Victor Anjos of 1000ml

Machine Generated Transcript via Descript

Dave Mathias: [00:00:01] Hey everyone. Welcome to another episode of Data, Able. This is Dave Mathias, and I'm excited to be talking with you today. Now give a little background before getting to the episode today. First of all, okay. This was recorded on a Friday afternoon. So realize maybe didn't have the highest amount of energy as I should have had. So apologize for that from the start at the same time, it's a topic that I am really passionate about and certainly a lot of things that I do are around, coaching and training and trying to do it, in an applied sense. And so one of the things that we talk a lot about our around this type of applied learning cause Jansen and Victor, they have a company up in Toronto and they focus more on the technical side of data teams and they do that applied learning while certainly at, Beyond the Data, we focus more on a more business side persons or when it's technical aside people, it tends to be more on the soft skills or success skills of those individuals. But again, always in that applied instance.

And so really we're both passionate on how education has been done and how it can be done better. So that's certainly one of those topics that we talk about. Uh, but love to get your take too. Cause a lot of people are undergoing different forms of education, whether it's through their kids, whether it's, something where they're experiencing because they're trying to love all up while you're at home. Love to hear what has been going great or what hasn't been going so great. And of course you can email me at dave@gobeyondthedata.com.

And let me know about that. Speaking of training, actually have some upcoming training and you can always check out all of our training at gobeyondthedata.eventbrite.com or just go to our website at gobeyondthedata.com. One of the things really excited about is the upcoming Data Storytelling Boot Camp which starts in October and it's a five week boot camp, part-time, cohort model that's interactive, but of course virtual. Because you know what, we're not going to be in person.

Okay enough about that though, getting into one of the things that, how I had met Jansen through a service called lunch club. And I wanted to let you know about this. Cause I think it's something you might benefit from in COVID times. So lunch club or their website is lunchclub.ai is a service where they match you up with like-minded folks, whether it's, you're looking for other folks in product or data or other, some other interests some community cause whatever it is. They'll try to match you up based on a few of the things that you have identified. So that there's good connection with that other person. You have a good conversation.

And, in that conversation it's a video conference. Every conversation has been fun and interesting, highly recommend it. And if you're looking for a way to get out there, check it out. It's lunchclub.ai. ,There's no paid promotion to do this . I had benefit out of it and I wanted to share with you. And I want to thank Jasmine RuKim, who is a former guest of the podcast, and she's the one who actually introduced me to it. So thank you, Jasmine for doing.

The other thing to mention before getting going today is this is actually a dual podcast recording because Janssen and Victor, they have their own podcast called All Things Data, and there's actually a couple of those out there. obviously this is the Data Able podcast. So we're doing a cross podcast that we both thought was a topic that we were both passionate about. And so we thought it would be an episode, that'd be good for both of our audiences. Do recommend you check out their All Things Data podcast, and subscribe to that. And we'll of course have the link in the show notes. And of course, thank you for listening to Data Able.

So now let's get into today's episode.

Jansen Sullivan: [00:03:53] Welcome to the All Things Data podcast combined edition. We're here actually with Dave Mathias, the data able podcast hosts. So we're combining our podcasts this week.

Dave Mathias: [00:04:02] Awesome. Great to be with you. we're socially distanced between Minneapolis and Toronto. So we're good there.

Jansen Sullivan: [00:04:09] Yeah, that's awesome.

No, thanks. Thanks for joining. And thanks for setting all this up too. I know it was a bit of back and forth, but, and we get to use some new podcast tech, so that's kinda cool.

Dave Mathias: [00:04:17] Yeah. It's always good fun. And talking about leading edge things, so we're going to talk a little data education, analytics, education, but tell us more about what you are about and, and I'll do the same and maybe go back and forth a little bit there and then go into the analytics education.

Victor Anjos: [00:04:32] 1000 ML really started as a data consultancy way back in the day. and as we kept going, we noticed very quickly that, All the people that we'd hire or work with and whatnot. they, it's not that they necessarily lack data literacy altogether, but they didn't have the calluses of a data professional, who's been in the world for various time.

And we're purporting to be, excellent data people, whatever that data people means. it could be data scientist or analyst or AI practitioner, whatever. And we kept creating, programs such that we could, it would take them from entry level emerging or what have you, and make them into fully fledged, fully fledged and fully rounded, uh, data practitioners that were super useful to us.

And very quickly, we found that a lot of our contemporaries in the community really wanted, some of our staff and wanted what we were doing. And it was a bit of a, of a moat that we were building around the business that allowed us to, scale and have a, I guess, Lower operational costs across the board.

So that led us to a world of let's create this for other people as well, and let's make the products, the people that we're training up. So that was the crux of a thousand of mouth. And we keep doing, consultancy work and project work and product work. Cause we have a, I guess, a dearth of data practitioners that come through and a lot of them are great that we want to keep them.

So we keep the ones that we really want around at times. when we have projects for them specifically, and if not, we help get them placed into jobs somewhere in the community. And now it's actually been worldwide. Really

Dave Mathias: [00:06:13] cool. That's a, almost like a sort of a apprenticeship like model, whereas the hand in hand, learning together, little old school, that's coming back more in popularity in some different cases, right.

Jansen Sullivan: [00:06:25] That's right. Well, and I think we're big.

Yeah, no, we're just saying we're big proponents of, hands on and, learning from textbooks is great. I think theoretically, but, uh, at the end of the day, employers and ourselves, we want people that can do things and actually operate and execute.

So we think that the fastest way to learn, especially in the data world, there's a lot it's trial and error. It's a lot of like experimentation and tinkering around. So we encourage people to do that and that's the way we deliver curriculum also.

Dave Mathias: [00:06:53] Cool. That's very cool. And somewhat, even though we both saw problems in the spaces we're in, we're sorta both saw some, problems in different areas, I think.

And so for us, we certainly, we focus on more the business side of data and the challenges that are there. So less the data scientists or folks like that, and more the product managers, the finance person, the marketing person, the HR person, and getting that data literacy up across the board, along with some of those, more success skills for the technical people, but we don't really get into the more rigorous training of more technical people, other than some of those success skills.

So I think we have some overlap that we also have, some differences where we're playing in

Victor Anjos: [00:07:38] that's for sure. And we, we keep noticing all across the board data literacy is very top of mind for a huge number of organizations. I don't know if it was a catalyzed really because of, the COVID and the pandemic and people working from home quite a bit and having to do more, maybe stress themselves a little more than they had in the past.

Uh, but yeah, like we, we keep running into conversations over and over and over again, with all kinds of organizations who are, uh, you know, basically telling us we are not literate enough and we really need to understand what kind of insights and what kind of power data really has.

Dave Mathias: [00:08:12] Yeah, I was in fact actually recording another data able episode just this morning, with these, chief data officer of a company out East that sort of like a medium sized software B2B type company and him and I were chatting.

And on this day of literacy of front is as one of the components and really the data translator role and data storytelling role. And he looks at it as there's going to be those folks that are uber technical and that are really good at it.

Yeah. It does sort of diving in and, being the rule really strong data scientists. And then there's this huge Number of people that are going to be, data storytellers, data, visualization, data literacy, and just with really strong domain expertise. how do you get more of that.

I just got off the phone with somebody that's looking to make a career change and she's done a lot of things. In fact, she was actually working for the Canadian Consulate at one point, here in Minneapolis and we're chatting on the first thing that she picked up, cause she was like, I want to leverage my creativity, but I also like the field of analytics and I was thinking about data storytelling and data visualization.

So the first thing she picked up was a Coursera Python course. And I was like, okay, I don't know if that's the first thing that you, why don't I dive into? I mean, that's great. Like if you're curious, but how do you, especially because you get a lot of folks that are career changers, right. And how do you.

Help educate and transition somebody that's maybe a career changing person where they weren't, let's say they want to jump into the very technical or maybe even more of an analyst or a data scientist versus like somebody that's coming out of school or they're pretty young in their career.

Jansen Sullivan: [00:09:44] Well, I think there's a couple of things, like first off, I think it's the hype of marketing. so the idea, everyone's saying like data scientists, Python, these are all the sexy things you need to be doing. And, sometimes they're just not, that's not the thing they need.

You know, and I think it's, you're, you're right by just kind of asking the questions and it's not necessarily coaching, but just kind of informing people that there are other roles outside of data scientists that are in the data world. But, you know, once you kind of get past that, I think a person who's been in industry for awhile career change, but it doesn't necessarily mean industry change.

Right? So these folks who've been in the game for a while and seen, the problems that, occur in their industry or at their business are extremely valuable. Right. I mean, it's like taking someone who's a great operator and then adding the data layer onto them. Right. So they know a lot of things that, you know, People like the general population don't know, so they can apply that kind of specialist lens.

And usually what I tell those folks are just like, what, or I'll ask them questions around the idea of like, what kind of problems do you see, or what kind of questions would you like to answer if you had all the data in the world type of thing, and then go try and solve that, you know?

Dave Mathias: [00:10:57] That makes perfect sense. I think people that often times. Want to do these career change. There is this article I'm still waiting to publish it. It's talking about the whole unicorn view of data scientist and product management.

I think both of those roles have this misconception of this is just like the roles, the two types of roles that are so sought after nowadays, other than being like the billionaire entrepreneur. Right. Uh, so I think that the question is really what's motivating. Why do you want to be that role?

What do you see your skillsets that line there? And like, what's your passion? Um, I think too many people are driven by the hype nowadays. And we think , oh, I can learn anything on a weekend or in a bootcamp. And certainly some boot camps can learn a lot, but that's only going to be like the very cursory level of your career as to be good at anything. It takes lots of time to be that polished, whatever it is.

Victor Anjos: [00:11:50] People are often drawn, just by the salary numbers that they see or have read in some magazines and publication, the economist, whatever it is. Right. And they're like, Oh wow. Data scientists make so much money that's exactly what I need to do. And it's also future-proofing me. So they jump in and think, well, I'm going to be a great data scientist, especially at the entry level where it's really tough to get hired as a data scientist.

Right out of school and it's really tough to get hired without the real experience. So oftentimes these people need to, think through, well, how do I get my foot in the door and do something that is very close to, or adjacent to data science and work my way into that data science workflow. So, as soon as I possibly can, so I demonstrate value and to give me more responsibility.

Jansen Sullivan: [00:12:36] I think it's also understanding the roles like data science is one of those things that, you know, it's one of those terms where you could just say like IT or developer, there's so many things. You could be in research. So you're just looking in models. You could be in the engineering side where you're productionizing, you could be in the analysis side.

Right. So I mean, it all depends what you want to know. And I think data science is this like catch all term, but. You know, we're seeing this now where you're getting new role titles or new specialization titles of like machine learning, engineers and data science, researcher, data, science analysts, things like that.

So I think, people don't know what they're getting into, so they're just like immediate, I'm taking Python. So they don't know the type of problems they want to be solving or like what they want to actually be doing. They just know that it's data is the new goal. Then I need to go mine for it.

Dave Mathias: [00:13:26] Yeah, that's a great point.

So for you guys, when you're talking with somebody initially that first time, what are the key questions that you're asking them to understand, okay, what is this person's motivations and how much do I want to invest in? Cause you're with an apprenticeship type model, that's a fairly significant effort from your end.

And also from there end. Their end and certainly if they're not doing great work, it's reflects on you, especially when you're doing like the consulting type work. Tell me a little bit of how that process goes.

Victor Anjos: [00:13:57] Well, generally to look at quality and ensure that we keep quality high, we generally work them through the model is we work them through, a ton of, projects all the way through this experiential learning, over a longer period than a bootcamp would, all the, while they're also being mentored.

So we have staff mentors who, sit by the wayside. So it's not a selfless program, but that would be pretty crazy. To try and get people to really do that. Um, that's more akin to a fellowship and that usually works or can work for, postgraduate degrees. But if you're, and I don't know, belittle it, but if you're a lowly, batch their degree, or even an honors bachelor, you are possibly not of the same rigor and critical thinking that, somebody with a postgraduate degree might have. So you need a little bit more hand holding and a little bit more showing you the way. So you were completely right at the start where you sort of juxtaposed it to an apprenticeship because that's really how we work.

It's the mentor is the one who is paid well and Israel charge of delivery along with an account manager for our projects. And we give it the rigor and professionalism that we do on the consulting side. It's just that much like, if you were talking about the Deloitte or Accenture or any of these big consulting houses, Um, you generally don't get the most expensive people doing all the work for you.

They're usually the network and they're the ones selling you the work rather than actually producing all the deliverables. They often will package it up and give it to you at the end. It does have a professional sheen to it, but. The, the majority of the work is often in source to, the, I guess the junior staff.

Jansen Sullivan: [00:15:42] Yeah. It's all about billability at that point, right. Where you're billing out at higher rates, but you're having juniors take on that, take on those roles. But I think in terms of like questions, like where the types of folks that we want. Was that the original question?

Dave Mathias: [00:15:56] Yeah. It was just when you're trying to identify is this a person that I want to take on at the beginning, because I think that's a big commitment on your end to go.

Right.

Jansen Sullivan: [00:16:04] And I think, the big thing is that the questions I am most, I guess concerned about like, you can pass a quick math test. You can pass those. We can test you technically, but it's the curious folks, it's the people who are wanting to dig, the people who aren't afraid to make mistakes.

The people that are, a little more, more, a little less risk adverse, I guess. Uh, but the people who are willing to try things out, like you can have a lot of practitioners, like, I mean, we're not looking for unicorns and we're not looking for. People who are necessarily like, great, great operators, great storytellers plus great technicians plus you can't always find that stuff.

So, we do look for people more about more around the idea of like curiosity, and people who are willing to learn. So, I mean, there's a lot of questions around that where you can just ask people how they would figure something out. No, just, I want to see how you think through problems. Um, and people who, people who want to, well, you can admit and just say like, things like, you know what I don't know, but I'll go find out versus trying to BS your way through something to make yourself like an authority.

But, it's, it's around the idea of just, uh, being an Explorer and like, Okay with okay. With, I'm not coming up with the right answer right away, because I mean, at the end of the day, data science and data analysis and technology and stuff, it's a lot of exploration. It's a lot of trial and error.

You don't know the answer up front a lot of the times. So you kind of go in half knowing and then you'll figure, you'll figure it out along the way. But I mean, you do have methodologies and frameworks that you work through, but it's not like a typical engineering problem where you know what you're going to do, you're just like, here's the problem, you know, I need to make code or I need to make a bridge or I need to, whatever. I just, these are the things, these are the constraints I work in. And then I just kind of build it up, you know?

Dave Mathias: [00:17:58] It makes lot of sense. And so one of the things I've seen a lot, I mean, certainly there's tons of Universities and we had chatted on this a little bit before where there's so many universities doing so many programs in this space, whether it's data science or analysts or engineering or whatever you want to call it, different programs and a lot at the master's level or, or boot camps, even things like that.

If you had the power to go into in that type of program, how would you change maybe one of your local universities and adapt their program? Would you try to do the whole program as an apprenticeship model? Or how would you sort of restructure, uh, universities, uh, education in the space?

Jansen Sullivan: [00:18:38] Well, I guess one thing is that, universities aren't really open to the idea of traditionally anyways, are open to the idea of like apprenticeship, it's very, everyone's going to say experiential learning, but I find a lot of like, some universities are doing this decently where they're having more of like applied or co-op programs and stuff, but actually to teach it, um, I think it's, they need to be taking more like case based learning, uh, Bringing in real problems and real data sets, not just like here's a public data set of, 5,000 records, perfectly formatted. So yeah, you have to do, you can do this piece analysis.

It's like, you have to understand the rigors of, doing these things and going through the pain of data. Um, What's the word, like, uh, the data engineering and then yeah, like the, the wrangling and stuff. So, and, but the thing is that should be a course because right now, the way university teaches is in discrete chunks, right.

You would take a data engineering course. Then you would take a data science course. Then you might take like a model course. And then, but at, in the real world, all of those are combined to solve a problem. Right. It's like taking a course in cooking, but instead of cooking, you just take a chopping course.

Then the next course you take is just a boiling course. And then the next course you take is a frying course. Like that's not how you cook. Right. You cook all, you have to take all the skills at once. So, that's my thought, like if you were to do something, it's more of like you might spend a semester solving problem and going in and working with, industry to do it.

How about yourself? I'm going to flip that back to you.

Dave Mathias: [00:20:16] Yeah. I would like to have a lot more experiential and I would, I think all learning well, so there's the master's level versus the bachelor's level. I would separate that out a little bit.

I think. A lot of the master's level learning that I've seen is so focused on checking boxes that are perceived to be important in organizations and checking as many boxes. It's sort of that quantity over the quality that I, get good at a couple of things and identify. What those people are passionate about what they're good at.

Um, like you said, we're not trying to create unicorns here. So I think for me, it would be spending a lot of time at the beginning to really know the student and then designing an education. That's going to be good for that person, him or her, that is going to have a significant experiential component.

I'm not against some of those courses that are gonna help get you along. But I think it's also the same thing as like, I'm not against doing, the hackathons or those types of things and like those data hackathons are great. They're good energy things, but it's not on point to a normal project that you're going to do, just like, if you want to do Kaggle competitions hey great, go do it. Um, but you don't have that interaction with the business and that asking questions and those types of things, if you're a data scientist or data analyst are such core features. So, bringing in real projects, the tough thing is, and in fact, one of the universities just was reach out to me saying, Hey, we're, we're looking for companies to do a part of this capstone. We had a couple of companies drop out and we need some companies to fill in and asked if I knew anyone. But it's the capstone project, right? It's the last thing that the students get to do. And it's like, Hey, if like, actually really apply it. Well, I think maybe that first semester there should be understanding where those gaps are for those individuals and what that person's good at, and then trying to accelerate them as they're doing multiple of those types of capstones throughout the the effort. And I also think the other question is on domain knowledge, if students are coming in with a significant amount of domain knowledge they're going into a master's program and they have 10 years of domain experience and in healthcare or an oil and gas or somewhere like that. That's a very different perspective and how you can lean in with that student and how education should be versus okay this person's just sorta very much on the path finding. I would question whether masters make sense for those students though too.

Jansen Sullivan: [00:22:30] yeah. And I think one thing is that they really understand, like when you've been in the workforce for a while, you're very, you're. You're more sensitive to your learning needs, right?

When you're coming out of school, you're just learning to learn. You're still in that kind of like head space where you're like, I'll learn anything. You know, it's not, it's not as targeted or focused. Um, but yeah, as an adult, like a more mature learner, you know, why you're there, you know what you're, you know, you're kind of taking stuff for the next steps for the most part.

Dave Mathias: [00:22:59] Yeah, that's a great point. So have you, you seen any one else doing so the apprenticeship model obviously has a long history in many different forms, and certainly there has been more of the emphasis on boot camps and things like that. But have you seen anyone else sort of taking on the charge, something similar to what you're doing in other areas of the world that you're feeling happy that things are progressing because the tough thing is scale, right? Like the model you're doing from a scale perspective, it's tough. A, a lot of people sort of buy into this type of approach and make things better.

Jansen Sullivan: [00:23:33] Um, we haven't seen anyone else doing this model, uh, in our. Well, not necessarily just in our backyard, we don't, we don't know anyone in Canada that's doing this, but we also don't know anyone really globally doing this. Uh, but there are folks who are doing other, um, doing this in other verticals, right? So, you know, there's sales and operations, um, people that are doing this for startups.

Uh there's I believe there's like a marketing one. That's similar. So, I think this model just lends itself to , the times right now people are looking, you know, if you're looking at it, job experience data science is not a first job. So if you see this as data scientists, there's always like the two to five year experience type of gap there.

So, or request even for junior. So how do you have a junior that has two to five years? That's not a junior anymore. I feel that data science like is being viewed as this like transitionary or transition type of career, where you started out as an analyst, then you might someone at your company gave you a shot and you move up into, data science, um, typically, but, scale is another issue, I guess like for us, we're looking to validate the model.

I think it's been validated, scale will be kind of the next problem that we tackle, but I don't think this model is, is not scalable. I mean, it's like boot camps, boot camps. I think we're at the same. Scale problems where it's you need instructors and you need, you know, mentors, like we call them mentors because we look at this as more as a mentorship program.

Um, and you still need people to deliver it. You know, this isn't a Coursera, like where you're just watching videos. I think that is a, that's a model for certain people, right? I mean, I think different learning methodologies are for different folks. I mean, some people need the in class, some people need a person, some people just want to do it on their own.

Right. Some people need books, some people need videos, some people need problems. So we're going after the people that want the experiential learning, not just like the classroom stuff.

Dave Mathias: [00:25:44] Is there anythings that you've seen out of COVID or other things recently that make you, excited or think that there might be finally a push to have better disruption in the education space?

Victor Anjos: [00:26:00] For me, I think it's largely been, uh, just sort of the, the backlash against universities generally. And I'm sure you've seen it as well. Like students are all up in arms as to why am I paying you this ridiculous tuition at this point? Uh, for nearly the same value. That I would get from some online learning platform, whatever that is.

Right. So I think there's a bit of disruption coming. I think the model has largely been flawed, not quite broken for quite a long time. Um, and in a world of capitalism, like it really has, it's a business. Let's go make money rather than, in a very altruistic way. Let's do the best for the student. Right.

So, I think it's very plausible that, we see a shakeup in it, but I don't know that the entrenched, colleges and universities are going to largely change unless they're losing money. So it's all a money game for them. And we'll have to see how this all shakes out in the next, I dunno, three to six or nine months, or however long there not allowing students into the halls of these higher learning institutions.

Dave Mathias: [00:27:06] so what are you most hopeful about with education that the data science or the analytics space?

Jansen Sullivan: [00:27:14] I think a big thing that we're seeing is just digital literacy. People are starting to really kind of put money into it. I, and people, companies are starting to put money into it. They're seeing it as not a competitive advantage anymore, but they're seeing that as table stakes, it's, it, I don't want to say, you know, alarm sound alarmist and saying like, this is do or die, but for companies to move forward and kind of be modern or modernized, uh, you need data, you need data to drive that way.

And that doesn't mean that, every one of your company needs to be a data scientist, but it's that ability to use data, to make decisions, ability to interpret data, the ability to form arguments with data, you know, the ability to discuss data. And it's not, I think it's inherent to everybody. I mean, a lot of people are like, Oh, I'm so bad with data.

I don't know this, that and the other. Um, but at the end of the day, I'm like, if you can look at your grocery or a seat and tell me, uh, how much did you spend on vegetables? You can do that. And you've just done data analysis, and it's one of those things where I think it just needs to be reframed.

And I think a lot of people are doing it every day. I think everyone's doing it every day. You just don't know that you're doing it and taking that in and applying it to the business context. Right. And just like how you're making decisions, how decisions are being formed. Right. Not just making the decision, but like, why am I making this decision?

And then getting into that idea of like evidence and using, any of that information to help you drive your business a little further. So I'm really excited that companies are investing into that or looking into it at least like it's. It's conversation now, at least. Um, and universities are pushing for this like machine, like the academic side of it, which is like machine learning. Um, you're seeing a lot more of these like master's degrees and analytics. Well, not, I mean, I don't know if they're good or not, but universities are jumping on the trend in that bandwagon and that it's pushing the profile and people are aware of it and people are talking about it.

So I think that's important, even just being able to have the conversation. Yeah. I'm sorry. I think it's a good thing. I think, if you're digitally native, we've talked about this on our own podcast, but if you're digitally native, this is an important, um, this is an important thing to look out when you're forming companies. And companies that are doing it right off the bat are growing at exponential rates. Like those are the companies that are killing it because they're able to use data right from the get go.

Dave Mathias: [00:29:50] That's a great point. And certainly. I think the sort of, "yes, and"ing that as I'm a big fan of improv is more leaders are even willing to start asking those questions and get a little bit better with data literacy.

And I'm surprised at how many leaders are, although getting to, I had, uh, I've had a couple of executives recently say, Hey, do you know a good Python class I could take? Which I'm always like. Okay. Like, tell me more, why is it just total mental curiosity, but yeah. Uh, but I think there's, there's at least just a broader recognition of the importance of this.

Like you say, across the board and taking a step back. Although, one thing I have wondered with COVID is. Cause you would think that, Hey, we're looking at more visualization and data and things like that just on a normal basis. I think from the general public and at least more journalists and things, or at least appear to be talking about it. But I do wonder, do you think COVID is a good representation of where that's, how pushed data literacy forward or do you think this is maybe some of the evidences of some of the shortcomings of data literacy in our society?

Jansen Sullivan: [00:31:03] Um, that's a really good question. I think, COVID it's interesting because there's data coming out about COVID all the time and it's being pushed to the public, but I don't think you can convince people who don't want to be convinced. So I mean, there's people who are looking at the numbers and interpreting them and trying to do things from them.

And then there's certain folks that know that numbers exist, but they don't care about them. You know? So I think that, um, The conversations around COVID for the folks that care about the numbers are starting to become more interesting. Um, I think that the, uh, I think that you can use that information to help educate the public and actually boost the profile of data literacy and inspire others.

And I think that inspiration comes. Especially at like the youth. I don't know if you can change adults' minds, you know what I mean? But the young and impressionable ones, um, can still do it and the ones that want to do something about it. But, um, I think it's not, I don't think it's necessarily boosting the profile, but it's giving you data to work with that's relevant and usable, and to have conversations about, does that make sense?

Dave Mathias: [00:32:23] Yeah, it certainly makes sense. But I do think most people are not in the camp that don't want to be convinced. I think oftentimes we think people don't want to be convinced, but really it's because we're maybe coming at them with a message presented in a form that seems to either be talking down to people or not really seeking their understanding, but basically dictating what the answer is.

So one of the things I think though, when we look at COVID, that really comes out is that there's just a lot of data and a lot of uncertainty. And even with the experts, you hear a lot of disagreement, uh, take place. And so understanding data and where that data is being, gotten, what that data really speaks to and what it doesn't speak to are challenging for experts, let alone the general public and certainly journalists and others. And so I think one of the things that this teaches us is:

a) we're certainly needing more data literacy than we have currently, not only by us, but also by our experts and

b) we really need to be thinking about. What type of data do we collect and how do we collect it? And what is that process and how is that communicated out to the public?

So if there's one lesson we can learn from this is to be more prepared ahead of time. And be used data ahead of time to be proactive instead of reactive, because we know from system one and system two thinking or thinking fast and slow, the research that Kahneman and Traversky did. That we aren't so great in the moment and making decisions and when our emotions get involved . And anytime you're having to clean up messes, instead of getting prepared for potential messes we know that you're not simply, it's just going to be a lot less expensive. And so there's a lot of stuff we could have done and we could do in the future, to minimize such pandemics and that from actually, basically shutting down our society for as long as it has or cause dramatic impacts even in areas of the world where it hasn't been as impactful. And so hopefully we learn from those lessons and we make changes and that's the best we can do uh certainly we can't change the past, but hopefully we'll learn from our mistakes and capitalize them in the future. Along with of course being better day literate professionals ourselves whether it's journals whether it's politicians or whether it's the general public

Jansen Sullivan: [00:35:11] I think it's a big thing with your leadership too. I mean, not your leadership being like the American leadership, but maybe that is a thing.

But, I think the important thing is that people showing people that, you know, To trust data, and showing people how to use data to come up to their own, come up to their own conclusions, but for people to be completely dismissive of it and say like, that's not true. It's like, well, what can you know like what are you going to say?

Like, you can't really say anything to folks like that, but I think, uh, you know, I'm a, I'm a big fan of American news and I watch, I watch it every day, but, uh, the interesting thing is, is that the current administration seems like they pick and choose the data that they want to use.

but then completely ignore everything else. So it doesn't, it's weird, right? Like they're saying don't listen to the data except the data I give you. So I think it's just having that ability to cut through what anyone is saying, look at the evidence and then ingest it and do it, come to your own conclusion about it.

But at least you can have a conversation with that.

Dave Mathias: [00:36:14] I think you make a good point when you're mentioning about really trying to find data to come to the conclusions that you want. And certainly this is a problem that we face in organizations all the time. And I don't think this is unique now or was in the past or will be in the future. I think this is just going to be something you're going to face and the more data literacy that people have, the better they'll be able to identify and be better at minimizing this occurring, but I don't think it's going to stop this from happening.

What are your thoughts Victor?

Victor Anjos: [00:36:51] I believe that a lot of the OG visualizers and I guess data scientists, they're not really scientists, but I guess, you know, largely visualizers, um, our data journalists, uh, whether or not they have the rigor of a scientist, that's to be determined.

But the news, the news media, um, for a very long time has always put this stuff together, uh, rightly or wrongly. And again, again, with a bias or a lens towards whatever they believe in a little bit more, uh, or whatever the, their bosses believe in, I guess, a little bit more. But that world, has existed and permeated within society for a very long time.

And, I feel like we've all missed an opportunity all throughout to really shine a light on them a bit and be like, this is some of the coolest stuff that can happen, uh, from the data world. And it's not necessarily going to be all AI all the time, all neural networks, all deep learning. You're not going to get into, AGI with everything you do, very often describing things really well.

And even at the level of counting things really well, uh, is a really good story to tell. And, people often get blinded by the sexiness of whatever the newest image recognition is or whatever the newest NLP model is when those are very hard things to achieve and get to, and you need an army of people, PhDs, infrastructure, all the stuff that goes around it to really do that kind of work.

Whereas if you're largely doing, really strong reporting and visualizations that tell a story and deliver business value. You're going to win and have a really good career. All throughout. Those are positions that are largely needed and hugely available as opposed to the data scientists that everybody believes.

Oh, well, if I go and do this thing with bootcamp X, then Google will come knocking or Facebook, AI will come knocking or somebody. And, people are kind of foolhardy about that, where realistically go and visualize really well. You'll have a great job.

Jansen Sullivan: [00:38:53] Yeah, that's right. Like, one of our friends, you know, to the program, Joseph he's been , working on is data visualization chops, and he's basically doing a viz a day in Tableau and some of his visuals now, like you look at what he started with and what he has now.

I'm like, man, just getting the reps in is really, has been really good, has been really, um, You know, through experience, it's giving him that ability to hone his craft. Right. But at the end of the day, it's a craft. Right. You've got to put, you've got to put the time in. You can't learn anything instantly, if you could, you'd be a professional athlete making billions of dollars.

Right. So, everyone needs more that's right. But , it takes time. Right. So I think a lot of people. Want to get take the shortcut, but at the end of the day, whatever job you do, you're a practitioner. And as a practitioner, you got to hone your craft. Right?

Dave Mathias: [00:39:42] Definitely.

You definitely have to put in the reps. If you're going to read one thing or look at one thing you've come across recently. Let's maybe answer this from the perspective of someone younger in their career. Whether it's a data scientist or data analyst or somebody more senior in their career, any recommendations, for either that junior, that senior person. This could be a book a podcast a course or whatever.

Victor Anjos: [00:40:06] Well definitely your podcasts for sure. I wouldn't get like overly bogged down, with a specific language, but like looking at, the kinds of things that are available, I pick up something like an introduction to statistical learning of some sort, like get your head wrapped around Stotts and maybe a little of Beysian learning.

Uh, and if you can start to understand that world a little more in, you already have the math background. Cause if you don't, it's a little harder to really gain that knowledge. Um, that's a really good crash course into a lot of the things that may. Be worthwhile for you to think of in the future. And then beyond that, like whether somebody chooses to do things in our Python or whatever, I mean, then it's chart your own path and kind of go down it.

Right.

Jansen Sullivan: [00:40:52] Yeah. I agree with that. And like, and I think it's on your background too. Like if you're, if you're, not super technical, maybe you don't want to do that kind of stuff. Like, you know, for the data science world, like I think everyone should learn stats and just. It's super helpful and it's going to help you in your job also, but there's some folks like, you could take the engineering route too, right?

Where you're going in you're less mathy and more like enj. So , along with if math isn't your jam, I think for the three of us, like math is our jam, so it's good. but if Math isn't your jam, maybe you want to go down the engineering road.

If you come from more of a dev background and like looking into it, Dev ops and, you know, maybe just setting up and learning, hunkering down and learning some AWS technology. Right? So like we're some cloud tech and like just getting good at like productionizing. Um, there's a ton of careers out there to in that.

So, I think it's where you come from, right? Like, I was talking to someone the other day and they, they come from a, uh, Risk and anti money laundering background. And they're like, Hey, I'm going to be, I'm taking a data science course. I'm like, cool. Do you like coding? They're like, no, I'm like, I don't what, it doesn't, it doesn't make sense.

You want to do this. And they're like, well, I want to be able to, create teams and talk to teams about what they're doing. And I was like, well, I don't know if you need to. I don't need it. I don't think you need to take a machine learning course to do that. Right. So maybe you need to, maybe you need to just read about it a lot more in terms of, I'm not saying not to take it, but I'm saying like, if you don't like coding and that's something you've always struggled with, maybe that's not, that's not the path, but I think a lot of people see it.

So I was like, maybe you need to learn more about AI strategy or get into more of like, The strats stuff or do some basic analysis first, don't go straight into, ML sometimes I think it's just a big jump because that's what everyone's talking about, but you need to, you need to run, you can't run before you walk.

Right. So I don't know. What do you think about that Victor? Like, is that a, is that crazy? Like, should I be telling people this? I feel like I feel that this is not advice life.

Victor Anjos: [00:42:59] No, it's smart advice. I mean, it's advice that you gain from having been in the field for a long enough time, right? So, people are eager and want to jump in and think that like, it's good that you think that you can do a lot of things and you can do everything.

Uh, but at times you sort of need the foundation before you build the house. Right.

Jansen Sullivan: [00:43:16] So that's right. I mean, do a visualization a day. Right. See if you like that stuff. And then, but I mean, I think, with machine learning and all this stuff, like the predictive side, it's like. You gotta be able to ask the questions, right.

And you can ask lots of really good questions, while doing reports, while doing visualizations and questions, we get questions. And eventually you move into those type of predictive questions, but that doesn't mean you have to be there right now, you know? Um, Dave, what about you? Like what, what do you think what's that what's that starter or that kickoff for you?

Dave Mathias: [00:43:43] Well, and, and so for that, that more junior person, I actually think getting just your critical thinking skills as much as possible. So there's a, I'm going to recommend a different podcast. Any other ours is Econ Talk and it's been going on since like 2007 and they cover a lot of different spaces. They do a lot of and certainly the word econs in it.

It's actually a former economics professor University of Chicago classically trained type of economist, but he is lots of different guests, whether it's Nassem Taleb or any of those types of guests that are on there covering a whole bunch of areas. And I think, understanding more the problems that different people face.

I think the more and part of it's like you get older, you get wiser. Right. But you see a lot of patterns. Like a lot of things are so similar, like. Industry to industry, problem people like people, problems, people problem. Like these things are just over and over again. Uh, it doesn't matter your role doesn't matter a domain.

so I think, but I do think, the more that you can fast track that and have that strong, critical thinking that uses data as a component of it, the better you're going to be in your career. So for me, Econ Talk has a been a good podcast to do that.

but on a perspective of one thing with people with like very basic programming skills, there's a fast.ai that Jeremy Howard does. And he has stuff around deep learning. He has two deep learning courses and he has a basic machine learning one and fast AI is a framework and it's also he does research around. This

Jansen Sullivan: [00:45:14] is great. It's just, it's so easy to implement.

Dave Mathias: [00:45:18] Yeah. It's so easy to implement, but the question is, is like, even like part of it's like, okay, start doing this stuff. Start seeing if you like it and what the job is. And like, cause I think sometimes people get so dissuaded cause they can't get very far, very quick and it's like, well, if you're a programmer that's wanting to switch into, or at least somebody that's got some basic programming skills that wants to make that switch fast. AI. Is great to give you exposure to say now what our granted, like, you can't just like set these things and like have no context around it. I know you have to go way beyond that to ever get any experience. And that's again, like doing applied learning is the really, the only way you can really do it well, I think I'm like, you guys do it, your folks. So, but I do think it's an easy thing where I put people down to say Hey given that background, this might be a good thing to you. Let's check in once a month. Let's see how you're doing it. Let's find some real projects that you might be able to find a nonprofit you can work with to help solve problem.

Now, oftentimes like the like deep learning, but you're not going to necessarily use, but you may be using some random forest to help a nonprofit in different capacities, for example.

Jansen Sullivan: [00:46:23] Right, right. Cool. No, that, that makes a lot of sense. And I think, just kicking it off, like, it depends on, it depends on the path, but really like, I think critical thinking is really the, a huge component of it.

And just how, um, you need to be answering questions, and critical thinking really helps you really helps you get there. So I think that's a great, that's kind of spot on.

Dave Mathias: [00:46:44] It's been a great discussion and it's Friday afternoon and we have to get somewhere right. But before we go where can the Data Able audience find you Jansen and Victor?

Jansen Sullivan: [00:46:55] you can find us on LinkedIn Janssen Sullivan and Victor's Victor and Joe's, so you can always find us there. Um, you can find us on the whole thing to say to podcast or on spot while we're off anchor, but, you know, a lot of our folks come off on Spotify. And I think, just partying thoughts.

It's okay. Uh, do things, get your hands dirty, find a project, and it doesn't have to be solving a problem a huge problem in the world. If you want to do a visualization a day that's, that's doing something. If you want to go help, help a not-for-profit you can, or if you just want to find, what NBA team is going to, whatever.

Yeah. The team's going to win the finals this year. that's a great project too, but find do something. And find something that you're passionate about and it'll keep you motivated. If you're doing something you don't want. You'll never, you'll never finish. No,

Dave Mathias: [00:47:37] exactly. That's great advice. And I just got off the phone with someone actually was saying, I was giving her advice for a visualization a week.

Now I need to car back and say every day,

Jansen Sullivan: [00:47:47] check it out. There's a challenge. There's a challenge. But, I can link you out to Joseph's, uh, visit day challenge, but I think a bunch of people are doing it, but it's, it's been really, it's been really insightful cause I've been following it on a.

I've been following it on LinkedIn. So sounds good.

Dave Mathias: [00:48:00] Thank you. And Victor, how about you? Uh, where do they find you? What's

Victor Anjos: [00:48:05] your. Uh, I'm on LinkedIn, Victor Anjos. Uh, I'm on Twitter, not super active. Um, but generally, LinkedIn is probably the best place cause it's very professional now.

Dave Mathias: [00:48:16] And same here . It's funny how we're all just referencing LinkedIn now. And I guess it's the LinkedIn's cool professional. So I'm DaveMathias1 with one is a number cause that's what was available for LinkedIn and I'm DaveMathias on Twitter, not Dave Mathias one. I was able to get the, without one although I'm not very active on it. So what does that say? Uh, but of course I gobeyondthedata.com is my company's website and Data Able on the, most of the podcast catchers that are out there. Great talk with you guys today.

Jansen Sullivan: [00:48:48] Yeah, no, it was great. Thanks for asking us lots of questions.

Cause usually it's just Vic and I asking each other questions. So it's good to have being a being on the other end of it.

Dave Mathias: [00:48:58] Inquisitive minds. Right. Well, good chatting with you guys. We'll do it again.

Jansen Sullivan: [00:49:02] Okay. Thank you. So cheers. Alright.

Learn more

Permalink

Ep 41 - Jeff Richardson - Data strategy and leadership from a B2B CDO

Beyond the Data — Thu, 27 Aug 2020 12:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“You’re going to have your data engineers and your data modeling on almost one side of the work and then your data visualization and your data translators.

They’ll sit in the middle and bridge that gap between your data engineers and modeling, and then your end users. People who need those insights to go make decisions.”

— Jeff Richardson

Being a Chief Data Officer is a challenging role but Jeff Richardson has been up to the challenge at Bentley Systems. In this episode Jeff talks about his approach around data strategy and leadership, the importance of data literacy and data translators, and more.

Jeff is a seasoned data and analytics executive leader with a cloud-first focused on evolving technology and trends. Over a 17-year career, he has crafted a results driven strategy for growth and delivered outcomes which have helped Bentley achieve a leading position in cloud technology, record revenue, and significant user growth. A prolific speaker on the topics of cloud and data and analytics, Richardson can often be found at conferences and networking events in the Greater Philadelphia and mid-Atlantic area (and now virtually!). He holds a bachelor’s degree from Providence College, where he was also a Division I swimmer, a master’s degree in Statistics from Central Connecticut State University and recently completed a business capstone program at Yale University.

More about Jeff

Bentley Systems: https://www.bentley.com/
Jeff’s LinkedIn: in/jeffrichardson

Transcript Download

Data Able Ep 41 – Data strategy and leadership lessons from a B2B CDO

Jeff Richardson, Chief Data Officer of Bentley Systems

Machine Generated Transcript via Descript

Dave Mathias: [00:00:00] Hey everyone. Welcome to another episode of Data Able Dave Mathias with you here today with a special guest Jeff Richardson. Welcome Jeff. Jeff is the Chief Data Officer of Bentley Systems. And Bentley is in the B2B space, doing software. And, Jeff will talk more about that.

but Jeff is a seasoned data and analytics executive leader with a cloud first focused, evolving technologies and trends. And boy, there's a lot of that happening more and more nowadays, right? Jeff.

Jeff Richardson: [00:00:30] Oh yeah. My God everywhere.

Dave Mathias: [00:00:32] And so Jeff, you want to tell a little bit more, give the audience some background because they see here to date officer they're excited.

They're like, Oh, that's how do I become a chief data officer? But how do you tell us a little bit about your journey to get where you're at today?

Jeff Richardson: [00:00:44] Sure. So since most people confuse Bentley with other companies with the same name, very quickly, Bentley systems is a CAD software provider. that makes software for building infrastructure assets. For anything in the world that you would think of that is a fit physical asset. So we are the leading provider of software solutions to people like engineers, architects, geospatial, engineers, that build things.

So if you've driven on a road in the United States, you've probably driven on a road that was built using software that we made. So at Bentley, my role of chief data officer is globally responsible for all of our data assets, our data architecture, and how we use information at the company. being a software company that sells.

Data sells software to users. Those users create data. We then take that data in the company. We do things with that. my role is a very broad spanning multifaceted, role in the organization. So I'm responsible for governance and architecture, but also monetization and insights analysis, even down to like corporate reporting and financial reporting.

Dave Mathias: [00:01:47] I was excited when I heard Bentley and I right away the, my mind went to the nice car. I'm sure a lot of people thought about that. And so thanks for the clarification now. Okay. So data governance and all those other things that you're responsible for, how do you. Keep the sanity between all the different demands that are on an organization now for using data.

how do you go about, just on your team to make sure that you're hitting on all the resources?

Jeff Richardson: [00:02:13] Yeah, so data governance is really an interesting topic right now. globally. So Bentley does business at 160 countries. So we have hit every single landmine of data governance in the world right now. We have a dedicated legal team that does compliance and Bentley as an organization that does business in Europe has a data protection officer.

And then in my team, I have dedicated resources who work with them on compliance. So it's very interesting. Most people don't realize the depth of the laws we've created around the world now. But if you do business, for instance, in Singapore, legally obligated to keep all of the data that you do business with Singapore residents.

In Singapore, physically, it can't leave the Singapore for boundaries. So we have to figure out ways to utilize data centers, Singapore, to hold Singaporean data on Singapore customers. Same thing in Sweden. A couple of years ago, Sweden had a huge data breach and their government, and they have now passed a law that any data on Swedish residents.

So if we sell software to a Swedish person or they're doing work on Swedish infrastructure assets that manage Swedish users, Swedish people, we have to keep that data in Sweden, physically in Sweden.

Dave Mathias: [00:03:19] That's a great insight. And I actually do wonder, because obviously we've had things like the California consumer privacy act, Where? Yeah. yeah. And so the question is, at some point. we run into those issues where people say, Hey, Yan, California residents. You need to keep your data in California.

And not that I've heard anyone. I haven't heard anyone talking about those things, but I think there is that real caution of, okay, this data is going everywhere. We need to figure out a way and it's sorta like a little hodgepodge. whether it's at the state or whether it's at the federal level, there's been discussions here in the United States.

What are your thoughts on that? do you have any perspective on where the direction of data privacy is going on? Data governance for US residents?

Jeff Richardson: [00:04:01] So for US residents were, we seem to always be a little bit behind other countries there. it's like the California laws are driving the US policy. I think. Globally my personal opinion on this is that what we're seeing now is a handful of reactionary measures to some serious events that need to be dealt with.

but globally, none of this scales very well, right? every large company now deals with data that's global, right? Everything that you do is it from a company that deals with things in many countries. So you can, there's no longterm play where. You can have these isolated country based laws or even state-based laws about information, Something as silly as Facebook as a global company, like. Anyone can deal with the Singapore and thus Swedish and the California laws, but when every country does that in the future, I don't think that's going to scale very well. but I find this to be a very interesting double-edged sort of discussion because it's forcing us now to take a more global approach to data.

So we're rolling out a new identity system in our company and the data laws in Europe and in Singapore specifically forced us to rethink how we architected that. So where we might've in the past. Stored most of that data in Eastern Europe, Eastern United States cloud public cloud vendor, like Azure or AWS, we now made decisions because of those laws to have it be globally scaled in multiple data centers, multiple regions with a more robust fail over system and DDR system and data residency systems.

So they drove innovation. Accidentally because of these policies. But if you take that any further, if there were six or 10 countries like that, I actually don't know how we would have built a system that managed to actually work for us following 10 countries where we had to be data resident. So the laws are driving a certain degree of innovation now accidentally.

but I do think eventually we're going to get to a point where there's just gotta be one, one set of rules. That's not maybe country specific.

Dave Mathias: [00:06:02] Yeah, that's a great point. And I do wonder always to where, especially when we get large companies, large technology companies that have the resources that can think about how to expand and have all playing all these different. Realms and in different ways, but if you're a startup or even an established company that might be, a few hundred people, so you that's, you've got a good point product out there, but you're working at you have it in a number of countries.

How do you, be able to have the systems in place to be able to handle that? And maybe there'll be. More companies that are just trying to do that service as if these laws stay the way. If the direction is that goes the same way, where it's becomes more fragmented, because I think that will be business opportunities for new companies there too.

But ideally we get to a point where there's at least a somewhat agreed upon sort of minimum viable standards of data. Privacy across the world and that'd be the ideal state, like you're saying. And I think in the long run that'd be great for innovation, but it's good that I was glad when GDPR was passed, Where it's like, Hey, it's challenge, companies to start thinking about this in a more holistic context of thinking of like, how do I treat data? And certainly with CCPA going then that was just another sort of, realm there. So yeah. When you go to that next level as a person that's in that chief data officer role, how do you work with, product managers and innovation people and other business leaders, as they're trying to, they hear data governance and sometimes they might want to think about, Oh, this is cutting off options, as opposed to really creating new opportunities.

How do you, as a chief data officer, be part of that innovation holistically in the company, as looking at new opportunities and partnering together, how does that work?

Jeff Richardson: [00:07:42] Yeah. So I'm fortunate in our organization where we developed the chief data role, and we saw a lot of these government requirements come up after we had structured process to develop products. So we fit in nicely into that process. More organically than I think it would have happened if it was the reverse.

So we generally have two kinds of development teams, Everything you can break up into two groups. We have the teams who come to us ahead of time and look for ways to architect and build those solutions that fit those roles and then work as a partnership to go forward and build that. And that works out great.

That is a great paradigm. I love that. And then of course there are the other side of that, where we have to step in. Nearly after the fact and kind of act as the data police. So we get that. Let me inside Bentley, we have dozens, if not hundreds of development teams working on some of our 700 different products.

So we get both sides of that a lot. We're seeing much more of the partnership. aspect the partnership paradigm happening because people don't want to have to go back and re-engineer these solutions afterwards, because it's no longer a debate. It's not Oh, we're giving you advice on how to do this.

we are telling you what the global laws are. We'll help you get there. We'll act as a trusted advisor and a partner and a resource for you to go do that. But you really have to go do this.

Dave Mathias: [00:08:59] Makes sense. And you're using technology that helps like snowflake, which is, great for being that flexibility. so doing more on that front too, I know we were chatting about, and so one of the things and partly is I'm speaking at a college. What's later today. it's to a group of business analysts and it's regarding the data translator role and how has business analysts, it can be better data translators.

What advice would you have for such an audience as a chief data officer that they could be better data translators, as BAS.

Jeff Richardson: [00:09:31] Yeah. So I really do think that the business analyst role and the data partnership role between like structured organizations and like the business teams is going to evolve quite a bit over time. And the value add spots are going to be the data translators and the data visualization is going forward, in those business roles.

So where, before you, might've had more centralized reporting and analytics and storytelling and translation. That we're seeing that certainly evolved now into the third or fourth or fifth iteration, whatever that is now. So business people, business analysts that can translate data models and complex data structures into insights and analysis are going to be huge in the next.

Two to five years. my advice to those people would be learn the tools of that translation, but don't get mired in the details of the data modeling anymore. I see that breaking up into two, two groups of people. Now

Jeff Richardson: [00:10:26] you're going to have your data engineers and your data modeling on almost one side of the work and then your data visualization and your data translators.

they'll sit in the middle and bridge that gap between your data engineers and modeling, and then your end users. People who need those insights to go make decisions.

Dave Mathias: [00:10:41] That's great. Yeah. and I, that same exact perspective. So that's good to hear. So at least I don't have to rewrite my talk either. That'd be a lot of stress. yeah. And so when you think about, in your organization, how do you as a leader of the whole organization and chief data officer.

how do you help empower that data translator role, that data storytelling role, is, people have the skills you're able to provide. Some of the expertise in your teams are able to work good from a technical team with those people. How, what do you do to, make things a better success?

New York?

Jeff Richardson: [00:11:14] Yeah. So just to be totally upfront on this, we are learning how to get to that now. So we're not there yet. We are in that transition between very structured, centralized data storage, reporting, and delivery in one team to distributing that out among many teams. So we're trying to leverage. You mentioned snowflake, Azure synapse, lots of tools to build systems that are scalable to lots of users to get lots of access points.

And then we're working on more structured and distributed training for those data warehouses, those data models and things, to give those to users, to go leverage in multiple ways and users today have. 10 and 20 and 50 times more resources to leverage those data models and data sources than we had five or 10 years ago.

it's amazing. the barrier for entry to get to that point has just completely fallen off. It's gone. So you can do insightful analysis. You can do deep dives into data. You can do very basic machine learning, even to the point of like advanced machine learning. Now with no barrier to get there, as long as you have good access to clean data.

So my team is focusing on building very reliable data pipelines that build very reliable. Data stores to share. And then we're less focused now on building those end points, the analysis, because it's almost impossible for us to figure out what analysis somebody is going to want. Very rarely does someone come to us anymore with a very structured list of KPIs and they say, go build these exact KPIs and we're not going to change them for five years.

It's here's the dataset I need access to. I really don't know what I'm going to do with it, but I've got 20 different questions I might go ask. And, I don't think you want to help me answer all 20 of those questions every day for the next 10 weeks. Let me go play with that with an analyst now who has, power BI and ClixSense and Tableau as a tool set to go build visualizations and maybe a little bit of R to go do some regression in some, general machine learning.

Dave Mathias: [00:13:17] Yeah, that makes a lot of sense. you're leaving that last mile to the business. The those and the flexibility, especially in this agile state. And so as part of this, one of the things, that's oftentimes referenced or more recently in the last few years is the concept of data literacy or data fluency.

And what is, I know you and I had chatted on this, before, too on, cause you have a fairly skilled, group of people given what type of work you guys do and a fairly data literate, CRA crowd, but you might even have that overly. educated, overly focused, a grip control. Look a little bit more like deep seeing what challenges you face in your organization as respects, add and little perspective for the audience.

Jeff Richardson: [00:13:56] Yeah. So I feel like data literacy has two parts to it. it's more of a data competence and then a data literacy in the strictest sense. So can you use the tools that you want to use to answer the questions and then inside those tools, do you understand the data sets that you're getting to? So data literacy.

Five years ago might have been. Do you understand how to use click view or reporting services to get your answers that you wanted? And now data literacy has many facets. I just did a talk with Jordan Morrow, the, the godfather of data literacy, about similar topics. And what we're focused on now is making sure that people understand the data lineage.

Of the information they get, where it comes from and how they get it. and the flow of the information, as well as the definitions then of the information they're getting. But to your point,

Jeff Richardson: [00:14:46] I love the phrase the last mile there, that last mile of analysis and that last mile of data literacy is really on the business analyst, the business user, the data analyst that's embedded in the various groups that are consuming this now to really take that to their team.

Jeff Richardson: [00:15:01] So we might give them revenue metrics or customer metrics or other usage. And in data points clearly defined clearly structured with how do you get this? What's the Providence of that information? as long as we give them clear definitions of that, I feel like all were onus of data literacy.

Got them there. They need to take that to the next level. We don't leave them in the wind to go do that by themselves. We help them also educate. They're users as well. but really data literacy has turned into a much broader topic. I feel just in the last two or three years than ever before.

Dave Mathias: [00:15:35] Yeah, that's great. And certainly Jordan's great. Like you're referencing, I have the ability to talk or opportunity to talk with Jordan many times, but also present with him at the connections conference a couple of years ago. good to hear that. and last time we were chatting, I think that same day you were, you had a virtual presentation with him somewhere,

Jeff Richardson: [00:15:53] it was like an hour after that answer.

Dave Mathias: [00:15:54] Hour. And after that. Yeah. so one of the things I think, when talking to people in your role that often in terms a challenge, challenges, like how do you continue to show value? As cause there's a lot of expenses in that analytic space. People see a very expensive software and very expensive people that have.

So the different titles, but to make sure that you're showing that your areas as providing value there, it's a, it's not just a breakeven, it's really can be a profit center, for the organization. So how do you go about that as a chief data

Jeff Richardson: [00:16:27] a great question. Great point. So when we chartered a data office and started the data team and built the chief data officer role, we built a bunch of tenants that we had that we wanted to fulfill in that role. And one of those tenants was a monetization of data. So it sounds very capitalistic and almost, yeah, negative, but we have all of this information, this data, and in the past we treated it like a thing that we were going to collect somewhere.

And as we've evolved, we've realized we have to treat that data as an asset. It becomes as important as our intellectual property and the physical things that we have that we run our business on. And other businesses that sell, we sell software. So we have no tangible good, but. Other companies that sell physical things are also learning this, that the data that they have in collect is becoming a real asset for their company.

And you don't just let an asset sit on a shelf. You figure out how to grow that asset. Either. Turn it into something that you can monetize and use to grow your company or something that you can use to appreciate the value of the asset itself. So again, one of the core tenants of our group was data monetization and turning that data asset into a larger asset.

We also focus on operationalizing things in the company. So reducing cost, reducing time spent in various places where we can operationalize data-driven tasks and help people reduce low value, add work, and give them opportunities to do more insightful work, more analysis. By taking away that silly burden of like data movement.

and then the last thing we do obviously is we were always trying to drive insight and analysis and give people new information to make better decisions.

Dave Mathias: [00:18:06] Of course. Yeah. Got to make those better decisions. And so one thing tied of this is. Is the valuation of data. So when using in the prioritization and just trying to think of like where to devote resources, do you do anything? And this is something most organizations are do, but, and we haven't chatted on this before.

It's do you do anything regarding that data evaluation, as you're trying to assess where efforts are spent or, and if so, how do you go about doing that?

Jeff Richardson: [00:18:32] so we absolutely do. So one of the data sets that we collect, is our usage data on software. So our data, our software model. Our business model is very similar to how you might do business with Microsoft Azure or AWS, where we invoice our users for the consumption of the products that they use sort of real time.

So it is a pay for what you use model. So we get streaming events of analytics, telemetry from this usage data. So the, we are constantly working to value and figure out ways to safeguard that information, and prioritize which information we get that we need to run our business on. So that becomes a critical payload of information.

We call it. And where that goes. And then we have obviously all the other information that we collect as a business, which becomes less critical to the day to day operations, but then might become critical to decision making and products. And then, those streams just keep breaking off into different value marked sort of sets of data.

Dave Mathias: [00:19:33] Great. That makes a lot of sense. And I know you and I, when we were chatting before your, you would assess your organization as a pretty upper end of sophistication of how data is being used in your organization. Certainly sounds that way. Is that fair?

Jeff Richardson: [00:19:47] Yeah. again, I've been very lucky in the role that I'm in the organization. We were a very, early adopter of cloud technology. So we sell software that we had hosted in our company to our users. So we were a cloud provider at some level in the past. and I want to say nine or 10 years ago, we very quickly jumped on the public cloud bandwagon.

we were one of the early adopters of many of Microsoft Azure services and AWS services. And we have continued to evolve that relationship as we grow. And we are always on the cutting edge of those cloud services, which is good and bad, obviously, because, Being on the cutting edge of any technology is always a bit dangerous, but, w we've been very fortunate to be a very cloud focused and cloud first organization.

Dave Mathias: [00:20:33] that sounds good. It sounds like you're happy. Guinea pig. Yeah, there you go. so one of the things I think is always a challenge is how, when you're working with other areas too, say between like analytics and business teams is how to make sure that everyone's getting credit together. And everyone wants to start to say, sales are up, things are up, it's out.

It's gotta be the B2B sales team or things like that. How do you work as a chief? Data officer in that realm of trying to, get, make sure that, you're getting credit for the stuff that you're able to add value, but at the same time, I'm making sure that the businesses getting credit when, with the wins too, is there, how do you, that sort of sharing of credit and in turn, like more resources that come about because of that?

Jeff Richardson: [00:21:20] That is a great question. and that is a constant sort of struggle, a discussion that we have throughout the organization, certainly with resources, because as you become more successful, In operationalizing, automating and streamlining work, it looks like you need less resources. And really you're just moving the work to different areas.

we're very fortunate again, in that we have lots of skilled technical people, so they don't assume that, the hidden machinery underneath the beautiful city, they live in, just works without people working on it. They know that underneath there are people who are getting their hands dirty.

Our business teams and our data teams really do work in concert and our executives are very good at understanding the visibility between how those groups work. So we generally try to partner with all of our business units and dedicate resources in those groups. So that we have a very visible partnership with them.

And the work that's being done is easily shared and easily separated between their groups and our groups. And we try not to speak in the terms of theirs and ours obviously, but you ended up doing that regardless. but by embedding resources in groups, and then. The opposite of that. We have resources in those groups that dotted line work through our data office.

It becomes more of a partnership and a shared service than it does, like a really hard it line where there's a fence and they throw requirements over the fence. by having this shared service sort of partnership mentality, we find that creates more of a synergy between those groups

Dave Mathias: [00:22:49] Yeah. Yeah, that's good. That's a nice type of perspective then. And so tied with that then. Cause then we were talking about data governance. We talked about some of the data privacy, but even data ethics. And where does the role of. How would data ethics be described? How would you describe it in your organization and as a person trying to really spearhead, that, what do you mean ethics and organization has?

How does that work when you're working with business and things like that? Because there's a, sometimes people will look at ethics as a hindrance or just like data privacy as a hindrance to innovation. how do you work with business to one is w. Maybe tell a little bit about what your data ethics practices are at your organization and also how you work with the business to make sure that, things are not just followed, but championed.

Jeff Richardson: [00:23:40] Absolutely. So we don't collect enough personal information to really fall on the spectrum where data ethics becomes is a serious issue for us. So again, yeah, we're mostly B2B and in the end users of our software, we know their first and last names and where they live and the software that they're using.

so it is some personal information, but we, it's not too far along the data ethics realm, certainly. When people are using our software to develop things, like roads and bridges or buildings or nuclear power plants, there is a day ethics discussion that happens, but that is all very locked down in the products.

We don't have visibility to that information. So the data ethics conversation stops there internally with our personal data. We've been leveraging global rules. And again, you'd mentioned, you were glad when GDPR came along, I am as well because does create a framework for ethics you can leverage and fall back on.

We've been able to leverage that for many different areas to create a. A reasonably good framework for an ethical way to deal with communication with users, collecting information on them, sharing that information and where people look at that as a hindrance or some kind of like speed bump along their road to, wonderful innovation.

We try to give them. Options and alternatives that give them equally beneficial, equally valuable information that doesn't violate those kinds of ethical gray areas on the edges there. so we've been trying to shift it to, yes, I know you want this, but here is an alternative to that. that's less of a gray area.

or when we have to, we do go back and fall back on the, these are the hard and fast rules. We are going to follow these rules. we will be the police of this. If we need to.

Dave Mathias: [00:25:19] Yeah. yeah. To have those hard guard rails in place. Yeah, you're trying to softly get people. one thing we had chat a little bit as COVID, obviously we're in Cova times right now chatting. but the world has changed a lot in, yeah. the responsiveness, the agile newness, pushing people had five, 10 years ahead in their businesses and how they're and certainly they're using your software.

How do you see the world with an, the B2B software space? How are you working to. Provide clients more value, and the chief data office as a chief data officer and your company in general, how do you see software being used differently now in this, COVID and post COVID realm?

Jeff Richardson: [00:25:59] Yeah, so we, we saw an immediate shift in how companies were handling their digital transformations. we provide software that helps, the giant, the largest EPC firms, engineering, architects, the people that build things, the constructors, we give them the software, they use to go build those things.

And we've been working with many of those companies over the last. 15 years on their various digital transformations, moving to the cloud, moving to more global mobility, moving to streamlined, edge devices and all the various things that you can bucket into the word salad of digital transformation.

COVID. Pushed many of those companies much faster into how do you get to true cloud where they didn't want resources stuck in their offices? They couldn't go to anymore. And how do you get true global mobility of their users? So if you have 10,000 architects and CAD designers around the world building.

Giant BIM models, multi gigabyte, multi terabyte, dim models. How do you access to that information when they are now working instead of out of 10 offices around the world, out of 10,000 homes with questionable internet access, questionable hardware, right? Questionable service all around. So we have been.

very fortunate to have software solutions to help them to go do that. And we are seeing a tremendous shift in how those companies want to deal with that. Many of them were saying, yeah, we know we'll go to the cloud in five years. in March 80% of those companies said we're going to the cloud in April.

Dave Mathias: [00:27:24] Yeah, it's amazing how that, both it's something where you probably been crossing your fingers, that this would happen and all of a sudden, whoops, now it's here. Yeah. Now you've got to deliver. yeah, so that's a good, that's a great thing. one thing we were talking about you in the chief data officer role, and I w.

I think that's something that's always helpful is to learn from our successes and our mistakes. If you're going to say, what was the thing that in your life you would say the biggest success that you had that contributed to you getting into the role that you have now? what is the, whether it's something that was totally locked and it just, happened or something that you consciously chose.

What was that thing that you would most attribute to? How, where you're at today?

Jeff Richardson: [00:28:05] so about 14 years ago, which seems like forever ago in real person time. I made a significant effort to formalize the way that we stored information in our company. and I built our first real. Internal accessible data warehouse for our colleagues. in the past data, warehouses had always been something that sat behind like a Cognos or a BW BI system that only experts could really get to.

And I really pushed to get, because at the time I wasn't in our it team, so I wouldn't have had access to that. So I really pushed to get that built for our company, for people that just have access to, and. Over the years, that's evolved into many other systems, many technologies and things, but that.

Kind of cemented my position as the person that brought information back to everybody in the company in a more open way. And it changed our culture of how we deal with information in the company. and I completely lucked into that because that is like the paradigm now that every company works off of.

And I just happened to have frustrations at the time that I wanted to get around. And I had read like an Inman book on data warehousing and I was like, Oh, we can do this and we can share this. And it was fine. And I just lucked into that a little early. And then built a name around that and then got to build on that into the role that I'm in now.

Dave Mathias: [00:29:25] Nice. and you're probably young enough into your career that you're like, whatever, like I'm going to push this. I'm frustrated and why not try to do it?

Jeff Richardson: [00:29:32] I had nothing to lose at the time and everything to gain. I, of course at the time, I didn't even realize that I just knew I was frustrated and wanted to solve a, so we just did that. And then we built a very small, I think it had about 20 objects in it, data warehouse, it had some customer data, some revenue, data, some data on materials and things, and people could answer questions.

All of a sudden that in the past was a big deal to go ask a question. And then it was just a little easier and then SQL became more ubiquitous as data tools became more ubiquitous. It just cascaded into this thing. And now we talk about our data warehouse almost. really, we talk about our data warehouse more than we talk about any other enterprise system in our company.

Dave Mathias: [00:30:11] Awesome. that's a great story. That's a great thing that people can learn from his second one is to take some risks when you're younger and understand when the upside is potentially upside. And two is just find a problem that really frustrates you and frustrates others. and maybe that's where you have put your reputation on, you won't necessarily know where it will lead you, but, oftentimes good things come about for that.

Jeff Richardson: [00:30:29] Very few people get mad at you for solving problems.

Dave Mathias: [00:30:32] Yes. Now the other side, the flip side of that is we all have failures like it or not. We were all we're. None of us are perfect, but, so what is the biggest failure? Whether it's something in school realm or whether it was in professional life or something like that you learn something from, that was a key component of your success now.

Jeff Richardson: [00:30:52] so I'd love to get into a funny anecdote on any of the other failures in my life, which are myriad and fantastic. but the first thing that comes to mind is our first foray into a really structured, modern, analytical visualization tool. so again, we had. Numerous tools in the past that were I T driven and sitting on those warehouses and were very structured and not very agile.

and when we rolled out our first BI tool, 10 years ago now, we had chosen ClickView as the tool that we were going to use. And at the time there was that big competitive soup of all of the other competitive tools at the time, Tableau power BI and the power pivot Microsoft suite was just coming out.

there was. The Cognos business objects things, and what we picked I'm very happy with. It was the right decision at the time. That was fantastic. and it's treated as well for years, but we should have done a better job of. Getting buy in and like ownership for more groups in our company, because what we immediately saw was, my group wants to use this and my group wants to use this.

And it created this useless argument and schism over who made the better bar chart, in a tool which. Just defeated the whole purpose of sharing information and getting that out. and at the time I wish I had been able to realize where we are now, where data modeling data storage, that pipelines of information becomes one part of that.

And the visualizations become much less important to control. You need to govern the data and how people understand the data and data literacy, but visualizations were going to become a commodity no matter what. And we're there now. So I wish I had known that then to just have a different sort of sentiment around how we did this, it would have just saved us a couple of years of, going back and forth over who made the better, Mico chart or whatever.

Dave Mathias: [00:32:47] do you think as part of the, is part of that lesson? Am I getting interpreted correctly as also bring those others into the table instead of just making decisions on something like this in a little bit more of a, less inclusive way, would you, is that one thing you're getting at there too,

Jeff Richardson: [00:33:03] Exactly. Yeah. Having ownership, not even just buying in, but ownership from all of the various groups, which is always hard, You're trying to get a whole bunch of ducks to swim in the same direction. but that would have been, a much better way to do that. And we've, I've seen this evolve in it over the last 10 years.

Now. It's much more common to get that ownership of other groups. And have them almost request of you the thing. And then you can implement that thing at the time. We knew what solved our problem best. And we were like, here's the best solution for you? And then immediately, of course, you're going to get someone who says no.

Dave Mathias: [00:33:34] Yes. I've seen this. This is better. It makes that better bar chart. So yeah. No, that's great advice. and certainly somebody like, no matter what level you're at, is something to think about. And how do you bring that? Get that buy in. From the start, I would say, early on yeah.

Conversations, don't bring people in late and early. and if don't want to be part of the conversation and then certainly, and you can still keep them updated. At least they felt like they were asked and contributed. yeah, there's a lot of ways to do it.

Jeff Richardson: [00:34:03] And even expanding on that. The other big takeaway that I've learned there is really try to look 10 years, five years, 10 years into the future and figure out what becomes a commodity and what becomes a core essence that you need to really focus on. And if I look back five, 10 years ago, I would have absolutely known that the Providence of data, right?

The accuracy of that information, that was never going to go. If we really had a core focus on that, but of course we were going to commoditize. The delivery of information, that last mile of information, there was no way that wasn't going to happen. So looking forward at what is the next thing that will be commoditized?

for instance, like I look right now where we are like, even like a year ago, machine learning algorithms and the languages of machine learning that is going to become a commodity over the next five years. So it doesn't make sense to try to push down. a hard structure on, you have to use R or you have to use, PSW, or you have to use some tool from machine learning.

That's all going to be commoditized. Every single tool is going to have regression and neural networks and all of that machine learning built into it. By default, the core there is going to be again, the data ownership and how you get that information into those tools.

Dave Mathias: [00:35:16] So as part of that is, okay. So data there's a, the proprietary data that you have versus all the more public data are data that's available to many. It might be vendors are providing it to many people, et cetera. how do you go about determining where to spend effort to build proprietary sort of data and modes, versus, just trying to leverage, partners with and with data.

Jeff Richardson: [00:35:42] We, I always try to bring in external data, into our sets to make them richer and to make them more useful. And honestly, we don't really have a good balance there as to how you justify what you do there. Cause the integration of that information into your systems is always the hard point. how do you make sure you match external lists or external research to your data in a way that you can join that and leverage that in some kind of model?

That makes sense. For the most part we have taken sets that we have found value in over the years and then cemented those into our culture of data. and that's worked really well. And we take the approach right now of taking the spaghetti and throwing it on the wall and seeing what sticks with other external sets of information.

And it, it hasn't been very structured and it hasn't been a great, repeatable process so far.

Dave Mathias: [00:36:33] Okay. So if you were saying the next couple of years in your role as CDO and where your organization's using data, what would you, where do you see the big efforts, the big wins happening? where what's the next two to three years look like?

Jeff Richardson: [00:36:46] So one of the things that I out of COVID I've taken out of this is going to be data sharing and data partnerships, and access to large stores of information that we use to run our business and that our users and customers use to run their business. Nobody wants to live in this world anymore, where they can't make very quick, real time decisions.

Everything is faster, it has to be reliable. It has to be automated. And what we're seeing with our customers is they want access to their stores of information. Very quickly and very reliably. And of course we do as well. We want that from our cloud providers. We want that from the people that we interact with on all kinds of levels to make decisions.

So we're seeing a much, Much more concerted push to automate and move and share information between organizations. We're going to leverage for some of that because they are, they provide us platform to securely share large amounts of information between ourselves and our vendors and our customers with their, I think they call it the marketplace and they may have rebranded it recently, but I'm hoping that we can leverage that in the future so that we can give large EPC firms access to the billions of records of data we have on their billable users and hours of what they'll eventually be charged for.

They want that. Now it's very hard to give them that reliably. So hopefully we can leverage tools more to do that in the future, less barriers between how you access information.

Dave Mathias: [00:38:05] Awesome. That sounds, yeah. Sounds very innovative and more, and certainly tools like snowflake are making that easier. When you now, one thing we talk a lot about is data storytelling. And as a chief data officer, you need to be able to tell a good data story. If you're looking at the data storytellers in your life, who's your favorite storyteller and who's your favorite data storyteller.

There may be different, or they may be the same person. so just storyteller versus data storyteller.

Jeff Richardson: [00:38:34] So my favorite storyteller out of like anyone who can just spin a tale, I would have to say is Simon Sinek.

Dave Mathias: [00:38:42] Sounds great.

Jeff Richardson: [00:38:43] his books, his the way he presents information, the way he tells stories, I just, I could listen to his podcast. I can listen to his videos, read his books for ever. I think behind me, It's two or three of his books as it is.

but as far as data storytelling goes, I usually fall back and default to a Mico Yuk.

Dave Mathias: [00:39:01] Yeah.

Jeff Richardson: [00:39:02] I love her podcast. I love the way she talks about showing people value, information and storytelling and the way she teaches storytelling to other people.

Dave Mathias: [00:39:11] Awesome.

Jeff Richardson: [00:39:12] I find a lot of value there.

Dave Mathias: [00:39:14] Great. Great. And of course I got asked this question, you referenced it when I'd sent it to you as okay. If you were going to be a data visualization, what kind of data visualization would you classify yourself as?

Jeff Richardson: [00:39:23] Oh, it's such a difficult question. There's so many, I think I would eventually end up with some kind of. Radar charts, something I like reaching out into lots of different areas and having like little points of expertise in different spots, but like a broad kind of view. And, also I've never been able to figure out a good use for a radar chart, which I think speaks well to like where I find myself now, like all over the place.

I'm trying to learn a lot, trying to figure out where I fit in different, situations and organizations. that's a great question. That's a crazy question. I love it.

Dave Mathias: [00:39:56] Okay. one thing we were chatting about is this week you've been working, you're taking this MIT course, and you it's around sort of operations. And it's got a lot of math around it and bottlenecks and trying to understand, some different things. But we were chatting on that a little bit.

What is the sort of the thing that you've taken the most out of that class and then maybe. As a tail add on question is what type of learning would you recommend to somebody if they're wanting to eventually be in that chief data officer role, where would you, what's something you had come across that you thought, Hey, this is somewhere you should go to take this class or read this book or do something like that.

Jeff Richardson: [00:40:32] Yeah. So the course that I'm in now on its operational excellence, I believe is the name and really it's about automation and how to streamline processes in any industry. So they focus on things that have inputs that have processes have outputs. and what I did not realize there is a very deep and well-researched science behind all of that.

There are. Structured mathematical formulas for how to deal with various different operational processes. in situations, there is a vendor, there is a mathematical function that will tell you how many newspapers to put in a newspaper stand, depending on a handful of them, of cost metrics. And that's it.

Formula is very well known in certain areas. I had no idea that there was actually math behind that, but there was really a news vendor, mathematical formula.

Dave Mathias: [00:41:19] yeah. One, one book. I like his algorithms and I forgot the author of it, but it's a great book. And it's, it talks about some of those different formulas in different contexts where you're like, Oh yeah, some things were like, okay, I know this, but some were brand new.

how about on a learning front? anything that you have come across. and it could be something where you wish you would've read this or taken this before you, years ago, but something you've come across that you would recommend to folks to learn. So that they'd be more successful if they wanted to be a chief data officer.

Jeff Richardson: [00:41:48] So specifically in my role, right? you're learning on so many different levels, at least I'm trying to, both business strategy, technology compliance. it's very. Broad. so places where you can go on now and get little vignettes of learning and lots of it areas like Pluralsight. that's been very valuable.

we were lucky enough to have Pluralsight logins through the organization I work for. So that becomes great. But what I've been finding myself doing more and more now, to plug your podcast and other podcasts is fine. Finding leaders in this area, leaders in the data, lytic space, the CDO space, listening to podcasts, and then writing down everything they reference.

And then reading those books, listening to those podcasts, watching those videos. So people like Cindy Howson, people like yourself, up in through people like Tim Ferriss, who like loosely is on the edges of data in some spots, figuring out what they talk about and then going and reading those books.

So algorithms to live by is going to be the next book that I go read.

Dave Mathias: [00:42:44] Hope you enjoy.

Jeff Richardson: [00:42:45] Yeah, I'm sure it's going to be fantastic. the last podcast that I listened to, you referenced the book, how to not be wrong, which is actually a very similar book. If I go look at the recommended things on Amazon there, those are right next to each other.

That was a great book on learning how to, how mathematics and structure around that can teach you ways to think about problems.

Dave Mathias: [00:43:05] Yeah. There's so many great books out there. one book that I read last year, did you have the chance to, Annie? Duke's a book thinking in bets.

Jeff Richardson: [00:43:12] I did not. Oh, she's the,

Dave Mathias: [00:43:13] player. Yeah.

Jeff Richardson: [00:43:14] player, the world

Dave Mathias: [00:43:15] yeah. And she's out of Philadelphia area too. So in fact, I met her last year when I was out in Philadelphia in person and she's working on her second book.

I don't think it's out yet. I think it's coming out this fall. but it's a really good book. And I think the more, one of the things that she talks about of course is like we, so it's a leveraging some of the concepts around behavioral science and. We're very, we tend to get stuck in our decisions.

And so thinking in bed says as really saying, okay, what do what, don't, how confident am I in these things? And what's the probability and you're almost never going to be a hundred. You should never be a hundred percent certain, but if you're 80% certain, you make the wrong bet.

Okay. how can I was my probabilities off, right? Okay, let's learn how to do that, but you don't feel so stuck to the decision. And so I think the more, I think the children, people with data, as they want to think that they can get to a answer that is going to be a hundred percent correct all the time and now, and I think.

Analytics will not get you there. It will get to a certain level of confidence and then you have to make decisions and you have to weigh a lot of trade offs. And that's the human part. That's not the, Hey, let's just stick it in an algorithm. And we'll just tell us how to like, run our lives on what we're done, So

Jeff Richardson: [00:44:21] down to decision science and the Condamine traverse

Dave Mathias: [00:44:25] yes. Yes.

Jeff Richardson: [00:44:26] the way to deal with how you think.

Dave Mathias: [00:44:28] Yes. Exactly. Exactly. so I think the more that data people can leverage those things in a part of it's like you're doing is you're grabbing information from all these sources to be better versed in that realm. So

Jeff Richardson: [00:44:38] that I've been doing that for a couple of years now. And it's really broadened, if you think about the spiderweb that you will learn from those kinds of things, right? you can figure out where to press on to go learn more things that are becoming more and more popular. ironically, I just listened to a podcast on graph databases, which is the next thing I'm going to go dig into.

And it really becomes like how graph databases work, where you constantly have over overlapping, nodes of things that intersect, right? Those become more and more prominent. Those become more important in your life. So the more and more you hear it, people that you respect talking about XYZ thing, that's probably the next big thing that you should go research.

Dave Mathias: [00:45:14] So on the graph database front, what makes you excited? I know you're just starting in that realm.

Jeff Richardson: [00:45:19] I had originally learned about these. We were out in Seattle about a year and a half ago at Microsoft headquarters talking about, how to architect a data solution for, a very distributed, set of alerts globally. and they had recommended, we look into a graph database to try to store.

That information to share it back as the repository of where it would store. So that was the first time I heard about that. and this was on my list of things to dig into more. And I just was searching for podcasts the other day, and I think it was on data crunch. they had a interview with, somebody who's writing a book that's coming out.

I think this week on graph databases. I very much think that will be in the forefront of data storage. Obviously many companies are using it now. Facebook's been using it for about five years now for their search. I think now as tools become more ubiquitous and it's less. the, again, that burden to get into that has dropped significantly. We're going to find that many companies, many organizations, many datasets lend themselves better to a graph structure than a relational structure. And in my head, I can think of five problems we have at Bentley that would immediately be easier if we could put them into, reasonably reliable, easy to turn on graph database systems.

Dave Mathias: [00:46:27] Cool. I know we're up at the bottom of the hour, love to, yeah. So love to have you share, how would somebody that listens to this podcast and wants to get ahold of you or follow what you're doing? Things like that. How would they do

Jeff Richardson: [00:46:41] Yep. So the best thing right now would be LinkedIn. I am working on building a social presence, find me on LinkedIn. just Jeff Richardson. I think my LinkedIn URL is just slash Jeff Richardson. message me connect. I love talking to people. this was fantastic talking to you is, was excellent.

so I love to network and talk to people about. Anything with data analytics or really anything in the general nerd space. She's watched the umbrella Academy recently. I just finished that. I would love to go talk to somebody about that now that I'm done.

Dave Mathias: [00:47:09] I'm up. So three right now on that, I'll be honest. And I started watching it because of your

Jeff Richardson: [00:47:13] That's right. We talked about that

Dave Mathias: [00:47:14] Yeah. Yeah. so because your recommendation, I was like, okay, I'm gonna start watching some, I just finished up so three. after I'm done watching, then we'll want to grab a chat. Okay.

have a good weekend ahead and a great talk with you and, till next time,

Jeff Richardson: [00:47:27] Thank you very much. Thanks a lot.

Permalink

It is time to put the BS in your organization, behavioral science that is

Beyond the Data — Thu, 11 Jun 2020 12:51:33 +0000

More and more organizations are moving to a product-aligned organization. This is a positive sign and a good step forward for organizations in their digital transformation. However, another needed transformation is taking form but is not as far along. This transformation is the integration of behavioral science practices and expertise into organizations.

Behavioral science is known by many names like behavioral economics, behavioral finance, behavioral psychology, people science, and others. No matter its name the definitions of these areas generally center around being "a branch of science that deals primarily with human action and often seeks to generalize about human behavior in society" ("behavioral science" per Merriam-Webster).

Some organizations like Google, Lemonade, and Walmart have even formed behavioral science teams and/or hired Chief Behavioral Officers. It is great to see the dedication of these companies. But, organizations don't need to hire a bunch of PhDs and develop a behavioral science center of excellence to start leveraging its power. In fact, it is our belief that behavioral science is one of those skills like analytics that most professionals should be versed on and have a certain level of literacy.

Before getting farther down the rabbit hole there are a few items worth mentioning as a starting point:

Behavioral science is indeed a science so it should be respected in that way by harnessing the scientific method. Yes that means forming a hypothesis, designing an experiment, running the experiment, analyzing the data, and iterating as needed.
Behavioral science is a powerful tool so it should be used for good, not evil. Google has the mantra "don't be evil" and that is something you should repeat and practice if leveraging behavioral science. This doesn't mean avoiding using it but maybe ask your customers and prospects how they feel with ways you are potentially leveraging behavioral science.
Behavioral science is very context-based so concepts sometimes apply strongly in some scenarios but not in other scenarios or in fact may apply in reverse. Plus, there are often multiple cognitive biases interacting in a given scenario so context matters. Hence the need to hypothesize and experiment as mentioned above.
Behavioral science is great for leveraging up what you are doing so look for ways to apply it today and start hypothesizing, testing, and implementing. You don't need a MS or PhD to get started but it is important to leverage knowledge of others when you can and later at the bottom of this article will be some suggested starting points.

Traditionally there had been the belief in fields like economics that most people were "rational actors" and this meant that people act what is in their logical best interest. With behavioral science there is a recognition that people are often irrational. In behavioral science the concept of "cognitive biases" are used to define systematic patterns of the deviation from rational judgment. This irrationality is often back to outdated biology that faced a different reality of worrying about being eaten at any moment and tight bands of tribes compared to our modern massive society where our predicament is too much choice, too much food, etc.

Now let's dive into some of the cognitive biases identified in behavioral science that that you should be aware about. This is just a sampling of cognitive bias but some important ones to get you started.

Confirmation Bias: Concept that people seek out information, evaluate information, or evaluate decisions based on their beliefs, thoughts, and assumptions.
Choice Overload: Concept that too many choices may result in indecision and associated unhappiness for a person.
Loss Aversion: Concept that losses are more impactful than gains. Or, in other words when people are faced with losses or equivalent gains, they are more motivated by the losses.
Priming: Technique where exposure to one stimulus influences a response to a subsequent stimulus, without conscious guidance or intention.
Status quo bias: Concept that people value prefer to be doing the same thing. This may be evidenced by not doing anything to change what is being done now or continued action to maintain what is being done now.
Sunk Cost Fallacy: Concept where people continue down a path or a decision due to the prior investment or effort that can't be recouped.
Time Discounting: Concept that people value items that are closer in time than farther off in time. Sometimes this discount is quite significant.

These are just a few of over 180 cognitive biases that are out there but even with these few you can start thinking about how you might take these ideas and potentially leverage them in your organization. For example, you might look at your communications and realize when you are messaging a loss and a gain in the same communication that the loss may have an over-weighted impression, so you might need to counter it somehow. Or, you might think that you might need to reduce the number of product choices being presented on your website at a time.

An important item to mention is that just because you are aware of cognitive biases doesn't mean that you can't fall victim to them yourself. For example, confirmation bias is something we all face even if you know what it is. But, there are indeed ways to counteract some of these biases. For confirmation bias one way its impact is to have diverse teams and another way is to ensure you have a good experimental design process.

Remember that you don't need to be a behavioral scientist by training or have a behavioral science department to get started and benefit. So if you are a product manager, human resource generalist, executive, data scientist, and many other positions it is time to ask yourself how you can better leverage behavioral science to provide your customers, employees, and shareholders a better experience and more value.

There will be future posts where we dive deeper on individual cognitive biases and other behavioral science concepts and applications to specific scenarios. In the meantime, if you want to learn more then here are some good places to start. Note there are many other great resources out there and this list is in no way exhaustive.

Websites: These are the few of the websites where I tend to access information on behavioral science.

Podcasts: As a person that is both a podcast fan and also a podcast host of Data Able, here are a few podcasts that either specifically focus on behavioral science or often incorporate into their episodes.

Books: There are many books in this space and here are just a few of the many that I would recommend.
In-person Groups: There might even be a behavioral science group in the area you live. Action Design Network is an organization that has groups in many cities but you might also look on MeetUp or just Google "behavioral science group in [city]" to find a group near you.

Happy experimenting and if you ever want to discuss how you might be able to harness behavioral science in your product, design, HR, sales, analytics, or other teams in your organization do not hesitate to contact us.

Need help?

Contact me

Ep 40 - Jasmine RuKim - Creativity Deserves a Better Track Record

Beyond the Data — Thu, 21 May 2020 10:30:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“Creativity deserves a better track record”

— Jasmine RuKim

Data is just as important for creatives and Jasmine RuKim as co-founder of Monicat Data knows all about it. Jasmine and her team help creative teams leverage data and technology to make a greater impact.

In addition to advising creative teams they are now in process of creating the Yellow Platform aimed at bringing data together for creative teams. Learn more about Monicat Data, Jasmine RuKim, and their IFundWomen campaign for the Yellow Data Platform.

More about Jasmine

Monicat Data: https://www.MonicatData.com
Yellow Data Platform Campaign: ifundwomen.com/projects/yellow-data-platform
Jasmine’s LinkedIn: linkedin.com/in/jasmine-rukim-mba-19461914
Jasmine’s Website jasminerukim.com
Favorite Storyteller: Prof G. www.profg.co

Permalink

Ep 39 - Brent Dykes - Data Storytelling is More Than Data Visualizatio

Beyond the Data — Mon, 06 Apr 2020 11:45:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“Data visualization is not data storytelling”

— Brent Dykes

Brent Dykes is longtime digital marketer and data storyteller that is passionate about data storytelling. By day Brent is Sr. Director of Data Strategy at Domo and at night he is an author both as a Forbes Contributor but also he authored the must have Effective Data Storytelling: How to Drive Change with Data, Narrative, and Visuals book. This is indeed a must-have book if you are involved in data whatsoever.

In this episode Brent and Dave both geek out on data storytelling but also on the principles around behavioral science that play an important role for data storytelling.

Brent believes strongly in the power of digital data and its ability to transform marketing organizations. He’s passionate about helping companies to become more data-driven and embracing the power of data storytelling. Brent believes that too many organizations and experts treat good data visualization as data storytelling but believes they are different. In fact, the differentiation of data visualization and data storytelling is in fact something Beyond the Data has taught and emphasized for a long time so it is good to see a book like Brent’s that focuses on how good data visualization empowers data storytelling but are not the same.

Brent is truly an interesting person. Prior to his current book he created PowerPointNinja.com blog and wrote Web Analytics Action Hero. In 2016, he received the Most Influential Contributor Award from the Digital Analytics Association (DAA).

If you like this episode you should also check out his recent talk at Domopalooza 2020 where he spoke on Effective Data Storytelling: The Crucial “Last Mile” of Analytics.

More about Brent

LinkedIn: in/BrentDykes
Twitter: https://twitter.com/AnalyticsHero
Effective Data Storytelling Website: https://www.effectivedatastorytelling.com/
Forbes: https://www.forbes.com/sites/brentdykes

Learn more

Permalink

Ep 38 - Allen Hillery - The Prolific Data Viz Writer

Beyond the Data — Tue, 10 Mar 2020 02:51:04 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“Data is everywhere and so it incorporates every facet of your life”

— Allen Hillery

Allen Hillery went from engineer to digital marketer to now a data visualization specialist not only being a prolific writer in the data visualization space but also teaching at Columbia University. Now he is taking his passion around data literacy and trying to make a difference in underprivileged young persons in New York City.

Make sure to check out this episode of Data Able and then continue on to read some of his engaging articles. If in NYC then maybe you will see Allen and NY Open Data or one of the other great data and community events.

More about Allen Hillery

Website: https://medium.com/@alglobehopper
LinkedIn: Allen Hillery
Twitter: https://twitter.com/AlDatavizguy
Publication: https://medium.com/nightingale
Highlighted articles: Three Reasons Why PowerPoint Gets a Bad Wrap & Three Reasons Why Storytelling is Important in Business

Permalink

Transforming Your Data into a Compelling Story

Beyond the Data — Thu, 20 Feb 2020 15:15:27 +0000

One question I often get asked teaching Storytelling with Data is, “What is the difference between data visualization and data storytelling?” This is a valid question. Industry-wide there has been a significant focus on perfecting and simplifying data visualization, yet much time hasn’t been spent on developing the storytelling process. It is presumptive to think that the presenter of the data visualization is wholly responsible for the story; in fact, the story starts building many steps before presentation.

Visual Data

The diagram above is based on Robert Kosara and Jock McKinlay’s view of the storytelling process. We can see that the main phases are exploring data, making a story and then communicating that story. This process is no way linear in practice but very iterative as shown in the sub-processes carried out by roles involved in transforming the data to a compelling story.

The data analyst usually spends a lot of time extracting, analyzing and summarizing a data set that will support the hypothesis of a story. This process can take many iterations as a story is being developed.

The scriptor builds the plot by visualizing the data to support one or more of the intended points the presenter wants to make. Most likely the analyst is also the scriptor but there are times where the business team is using the excerpts to tell their story.

The editor prepares the story material. Again this can be a shared role the scriptor takes on but it can also be the person in charge of laying out the presentation in either a slide deck, infographic or video.

The presenter delivers the story. This can be the executive who is relaying insights to their management team.

Storytelling can be a broad term. According to an article published by IEEE, “Most descriptions of storytelling require some sort of controlled delivery or presentation of information.” I recall many times earlier in my career being asked to pull data on sales volume and summarize by month and product category and pass the spreadsheet along to our sales team. The process for me ended with the explore and analyze phase. Over time these types of requests have turned into graphing those trends to adding a few lines to summarize why these trends are being observed. We were beginning to order the story.

Getting back to our question on the difference between data visualization and data storytelling, the breakdown in the process happens between delivering excerpts and ordering the story pieces. Let’s walk through an example. First I pulled data on temperature trends for the first day of Fall in NYC for the last 5 years. I summarized the results below.

The story I’m looking to communicate is how NYC is experiencing warmer than normal temperatures during fall. Our first official day of autumn 2019 was 15 degrees higher than the normal temperature of 73 degrees. I wanted to see how we’ve been trending over the past 5 years. While the data table gives me an idea temperatures have been increasing over time, visualizing the trends will make it easier to see.

Looking at the visualization above, we’re starting to craft a story. The first thing that allows this data visualization to cross into the realm of data story is it’s title, “First Day of Fall in NYC has been above average for the past five years.” This title begins to emphasize the message of the graph. Adding additional things like adding an annotation expressing that fall temperatures peaked at 89 degrees begins to support our story. A written narrative can begin to transform this data into a compelling story continuing to emphasize our message. Here’s an example - “NYC has been experiencing warmer than average temperatures for the past five years peaking close to 90 degrees this year! That’s 15 degrees higher than average!”

We can continue to build on this story by using these facts to support how this impacts fall foliage in the area as we know that temperature is a key factor in the vibrant hues we witness during this time. What I like about Kosara and McKinlay’s diagram is that it reminds us that data storytelling doesn’t stop at data visualization. It’s an iterative and collaborative process that has better success when feedback is given and taken to refine the story. You might end up being the analyst, scriptor, editor and presenter. If possible, get feedback from your peers if your manager doesn’t have the time. I believe as time goes on, we will begin to see visualization tools have a more robust story mode where it will suggest data snapshots or highlights to incorporate into your story. Imagine if it could order those story pieces into a slide show? That would begin to balance the scales of data viz and storytelling!

About the Author

Allen Hillery

Adjunct Professor at Columbia University,
Writer and Editor at Nightingale, a Medium.com Publication

Allen serves as part time faculty at Columbia University’s Applied Analytics program. He has extensive experience in developing and executing data analysis and integrating results into marketing programs and executive presentations. Allen is very passionate about data literacy and curates an article series that focuses on the importance of creating data narratives and spotlighting notable figures on how their use of storytelling made major impacts on society.

Follow Allen:

Website: https://medium.com/@alglobehopper
LinkedIn: Allen Hillery
Twitter: https://twitter.com/AlDatavizguy
Publication: https://medium.com/nightingale
Highlighted articles: Three Reasons Why PowerPoint Gets a Bad Wrap & Three Reasons Why Storytelling is Important in Business

Harnessing systems thinking to better understand and impact change

Beyond the Data — Tue, 18 Feb 2020 13:00:00 +0000

The world is highly complicated, and most things are made up of many things that together work as a system, whether it is the human body, your car, or even your city. Most things around us are systems and have a lot of moving parts and processes that are interconnected. This also includes most things you will encounter, whether you are a data scientist, product manager, customer experience architect, product designer, or leader, to name a few.

There is even a whole field of study called systems thinking to help us understand, design, and innovate these complex systems. Systems thinking is "a set of synergistic analytic skills used to improve the capability of identifying and understanding systems, predicting their behaviors, and devising modifications to them to produce desired effects. These skills work together as a system." Ross D. Arnold and Jon P. Wade / Procedia Computer Science 44 ( 2015 ) 669 – 678.

Systems thinking is a skill that every data and product person should have. It is vital to identify and understand systems whether you are looking to identify and solve customer problems. This is important because systems are impacted and react to change, so when you affect something with your desired change, something else is changed. A Systems Thinking approach will allow you to understand the system holistically and anticipate and look for those impacts. The better you are at systems thinking, the better you will be at affecting organizational change.

One place to start applying systems thinking is with your metrics and key performance indicators (KPIs). Each metric and KPI has an impact on your organizational system. Understanding and mapping out the metrics from your organization, from your strategic objective metrics down to each department and team's metrics and eventually down to individual metrics, will give you a holistic view of the organization from a metrics perspective. This will help you understand where friction exists, whether intended or not. This will help you know where strengths and weaknesses exist.

Another place where systems thinking plays an important role is in the area of personalization. Personalization is applying personalized service and offerings to your customer, whether you are a retailer, healthcare provider, financial service provider, etc. When you are seeking to provide personalization, you have a complex system of interactions with your customers that can generate significant amounts of data. Whether changing how you personalize marketing, in-service experience, recommendations, etc., you will be impacting your system. The more you can understand the overall system concerning the customer, the better you will be able to adapt resources to affect positive customer change and anticipate bad outcomes resulting from proposed changes.

The more you look around your work and world, the more I hope you start seeing systems. Much of our world is a system of systems. This allows us to zoom in and out of things and get perspectives of things at a more detailed and macro level through this zooming in and out. Take the opportunity to think about some of the systems you are part of in your organization, community, and personal life.

Let's say you have bought into the importance of systems thinking, and now you might ask, what are some things you can do to enhance your systems thinking skills? Here is a list of a variety of different resources you might want to check out:

I hope this post has motivated you to understand the value of systems thinking, but I hope it has inspired you to enhance and apply your systems thinking skills.

Need help?

Contact me

Ep 37 - Rosa Pantoja - Evolution from a Data Champion of One

Beyond the Data — Tue, 28 Jan 2020 12:06:46 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“The most important skill for an analyst is storytelling.”

— Rosa Pantoja

Rosa Pantoja started as a Data Champion of one and through her career brought others along to ride the wave of data. She is happy to play devil’s advocate and stir the pot but is always seeking to move the conversation forward to get to adding business value.

With over 20 years of experience you will learn from Rosa’s breadth of experience and practical application of bringing people along for the data ride.

Check out the whole episode for lots of great tips and inspiring stories from a true Data Champion!

More about Rosa Pantoja

LinkedIn - in/rosapantoja
Website - www.rosapantoja.com

Permalink

Data Viz Solves a Contradiction

Beyond the Data — Sun, 26 Jan 2020 19:52:00 +0000

The takeaway of this post is that good data visualization or data viz for short helps bridge the contradiction of fast impulsive thinking and slow deliberate rational thinking. It allows us to harness the power of data and rational thought in a fast-thinking manner. Up and to the right has a meaning. Red has a meaning. Line charts have a meaning. If using these and other data viz artifacts well then we possibly solve or at least reduce the thinking fast and slow contradiction as respects incorporating data into decisions.

Now if you have a little time here is a little more context…

One field that fascinates me is behavioral science. I think it is vital for both data and product people to understand and utilize. For those unfamiliar with behavioral science it is defined as “a branch of science (such as psychology, sociology, or anthropology) that deals primarily with human action and often seeks to generalize about human behavior in society” by Merriam Webster. There are other terms people use when referencing this area like behavioral economics, behavioral finance, applied economics, and others but if not the same they are close cousins.

The reason why I am bringing up behavioral science is because there is a segment of behavioral science which Daniel Kahneman, PhD, won the Noble Price in 2002 for the research where he applied psychological insights to economic theory in the area of judgment and decision-making and uncertainty and had done in tandem with the late Amos Tversky, PhD. Later he wrote the book around this research and its findings called Thinking Fast and Slow.

This is a lot of lead up but if you got this far great. A concept identified in Kahneman and Tversky’s research was around fast natured thinking representing people’s thought and decisions that are instinctual, gut-based, or in-the-moment and it is identified as System-1 Thinking. On the other hand, there is slow thinking which represents people’s thought and decisions that are rational, deliberate, and thoughtful and it is identified as System-2 Thinking. The premise is that each of these are valuable and play a role. Obviously if a lion is chasing us, we want System-1 Thinking but same is true if a car pulls out in front of us. We also we want System-2 Thinking when we want to identify shelter or where to drink in our past but same is true nowadays when we make career, financial, and family many family decisions. The point is these are both valuable and needed. Stress that every person should read Thinking Fast and Slow or at least read a more extensive summary of the research and outcomes than described here.

Reason why I am bringing up behavioral science and then this concept of fast and slow thinking is as data people we often use data visualization to help with our messaging and relaying information to others. The beauty of data visualization or data viz is that it is something that can solve the contradiction of opposing forces as my friend David Quimby would say. I think that data viz can do just this allowing fast thinking using with deliberate data information input.

Data viz allows us to help end users that may be prone to fast or instinctual to harness the power of data to leverage data in the decision process. However, one important part of this is using data viz in a way that is both using best practices, well understood concept, and done in a way that shows the data in an ethical manner.

Recently I led a data viz and storytelling workshop where someone that worked in data for a large healthcare provider attended. In our discussions she indicated that her organization decided that red could no longer be used in any charts, graphs or reports. The healthcare provider believed that using red led doctors to getting tunnel vision given that doctors are so focused on fixing problems once identified that everything else displayed would be ignored. This is a very astute position for her organization.

Of course we ideally do not want people to make fast or gut-based decisions when data is available and there is even a little time to decipher information. But, to press against or even augment these fast or instinctual decisions then harnessing good data viz is a manner to do so.

Learn more about behavioral science:

Podcasts: Hidden Brain, Freakonomics, and Behavioral Grooves
Blogs: Dan Ariely’s blog, BehavioralEconomics.com, and INudgeYou.com
Books: Freakonomics, Predictably Irrational, and Nudge

- Dave Mathias

Follow me on LinkedIn

Learn more

Ep 36 - Susan Otten - Convergence is Coming

Beyond the Data — Mon, 20 Jan 2020 13:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

This week Dave talks with Susan Otten of the Minnesota Chapter of the International Association of Business Communicators and founder of Indie Do Good. They talk about the upcoming importance of data for communications professionals and the upcoming Convergence Summit 2020 in Minneapolis, MN on March 19th where Dave will be speaking about "How to Use Data to Listen and Communicate Better."

More about Minnesota Chapter of IABC

Chapter Website: https://iabcmn.com/
Convergence Summit 2020: https://summit.iabcmn.com/

More about Susan Otten

LinkedIn: https://www.linkedin.com/in/sueotten
Website: IndieDoGood.com

Learn more

Permalink

The Intersection of Data Literacy and Data Governance

Beyond the Data — Tue, 14 Jan 2020 13:00:00 +0000

Data Governance and Data Literacy are two sides of the same coin. They both are about helping an organization maximize value from data. They both are about helping an organization broaden access to data. They both require a lot of change management and adoption at a top-down and bottoms-up level.

Now let’s dive a little bit more what is Data Governance and why it is important for Data Literacy efforts. Data Governance has many definitions but our view is that it is the systems, policies, and procedures where organizations manage access and availability of its data assets. While Data Literacy is the ability to access, understand, and communicate with data to make better decisions.

Accordingly, Data Governance is important because without policies and procedures around managing access and availability of data then there will likely be misunderstanding and miscommunication and, in turn, friction on accessing data by data literate users. This will demotivate those “data literate” people to often search for opportunities elsewhere.

One thing we have suggested in the past is to establish your Data Principles. Your Data Principles should contain your high-level operating standards around data including those related to Data Governance. Are you going to be a trusting organization internally? Are you going to be highly-restricted Data Culture? Are you going to capture as much data as you can? Or, are you going to be deliberate and limited in your data collection efforts? Whatever your Data Principles are they must be bought into and carried out whether specifically written down or understood. Bought into and carried out meaning that policies, procedures, and incentives are aligned to carrying out those Data Principles from leadership on down.

Certainly regulations both domestically and internationally have a strong impact on Data Governance minimum thresholds. But, at same time we would challenge you not to look at the minimums to establish your Data Principles. Determine the culture you desire and have your Data Principles reflect these for Data Governance. Then, as part of your Data Literacy efforts ensure that everyone is aware of your position on Data Governance and your Data Principles.

We will get into future posts around culture change, but your Data Literacy training should align with your Data Principles and, in turn, your Data Governance policies and procedures. For example, if you are going to maintain a very siloed organization then adapt your Data Literacy training to take that into account. Alternatively, your organization may believe in an open innovation culture and accordingly your Data Literacy principles should take that into account. Organizations of all shapes and sizes can benefit from Data Literacy, but aligning your training to your data principles and Data Governance will make it that much more effective and at same time strengthen your organization’s Data Culture.

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

Getting started on your right foot in metrics

Beyond the Data — Thu, 02 Jan 2020 15:50:00 +0000

Getting started off on the right foot in metrics requires that you start with a strong upfront effort in researching metrics to be implemented. This is true whether it is an organization-wide metric, team metric, or individual metric.

Good metrics start with understanding and defining the purpose. Understanding the purpose requires you to ask what you are trying to accomplish by creating, managing, analyzing, and communicating your specific metric. It is also important to realize that oftentimes, multiple purposes are served by metrics.

Here are the seven different purposes metrics serve for us:

De/incentivize behavior: Metrics drive behavior. If you’re tracking the length of calls in a phone center, it may incentivize ending customer calls too soon. Pay attention to unintended behaviors. It is important not to assume people are rational. Having a way to monitor outcomes while at the same time not being creepy is ideal.
Reduce uncertainty: Metrics help reduce uncertainty by ensuring that everyone is on the same page in the same way. Making decisions gets easier if everyone knows what’s going on. Reducing uncertainty, of course, is not removing uncertainty. Part of the metrics around reducing uncertainty should also be around helping understand uncertainty.
Show progress: Metrics are often used to help show progress on something. Perhaps It’s a sales target, or tracking marketing spend, or completed tasks on a project. One important thing about showing progress metrics is being aware of repercussions, both positive and negative, of this progress tracking and display.
Communicate priorities: Metrics can be used by leaders to communicate priorities and showcase key areas of focus for their organizations. Simply showing someone that you’re monitoring the situation can help provide focus. With that said, when creating new or updating metrics, it is important to have a strong communication plan and not just leave the metrics themselves to communicate.
Define expectations: Metrics help guide team members in your organization and help provide a mechanism for driving decisions and actions for individuals, teams, and organizations. As mentioned above, communication in the rollout of new and updated metrics is critical, and this becomes extra important for people where metrics are there to define expectations.
Inform others: Metrics help inform others and keep them in the loop. It is important to provide context regarding metrics to ensure well-informed metrics consumers. For example, this may be a sales management dashboard to inform of current sales status.
Create alignment: Metrics are often used to help align people, goals, marketplaces, etc. The bigger the organization, the more challenging and important creating alignment as a metrics purpose is. Metrics with a key goal of creating alignment should be prepared to update and re-update metrics, as initial efforts are often far from perfect.

Understanding the purpose of metrics requires you to know the audience for the metric, including who is impacted and why, along with who and why wants to do the impacting. Further, understanding the purpose should include ensuring alignment with the organization's needs and strategies.

I have compiled a Designing Metrics That Work Quick Start Guide to share some of our experiences over the years. Our passion for this is deep, and I even conduct public workshops around designing better metrics and, of course, offer the same directly to organizations. The world and organizations are constantly in flux, and well-designed and well-implemented metrics are key to maximizing success.

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

Ep 35 - Angela Wilkins - What Real Life Data Science Looks Like

Beyond the Data — Wed, 23 Oct 2019 04:47:27 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

This week, we talk with Angela Wilkins from Mercury Data Science. She has an amazing background in Data Science and has built a team to help startups leverage their data. She tells some amazing stories about what her data projects have delivered and how she went about solving them.

More about Angela Wilkins

LinkedIn - in/adwilkins

Website - www.mercuryds.com

Permalink

Harness the power of data visualization in products

Beyond the Data — Tue, 22 Oct 2019 11:00:00 +0000

Data visualization or data viz has become commonplace, and more and more products have incorporated it and analytics into their apps. This article explores how you can better implement data visualization as a product manager or product designer, which will hopefully lead to greater product adoption, product use, and product value from your users. We will go through some of the good, bad, and ugly of data viz in product and some steps you can use to ensure yours is good.

First off, let’s talk about what we mean by data viz in product. In product could mean actual data viz in the product, but it could also be something that is used in marketing, support, or other components where the end user of the product uses the data viz. Further, data viz doesn’t mean that it needs to be digital data viz with interaction but simply could mean data viz that is used in a cardboard display or a product sheet.

Secondly, let’s take a quick step back on what data viz is and why data viz matters. We define data, viz., as the representation of information in a graphic or visual form, often in the form of charts, graphs, tables, pictures, etc. Our goal is that any data viz should seek to have a defined user(s) and defined objective(s). The objective may be to inform the user or get the user to do or not do a certain action.

Before getting into the mini-dissection of examples of what we think was done well and what opportunities exist, let's get into a framework if you are in the product and looking to incorporate data viz. There are six things that we think every product person, whether a product manager, product owner, user experience, or product marketer, should think about:

What is the customer need seeking to fulfill with the data viz? If there is no customer need or pain point seeking to be filled, then there is no reason to include data viz. Remember, if you are not providing your customer value with data viz, then why are you using it? This may seem obvious, but I think even from our examples below, you will still see that this item can be missed.
Does satisfying the identified customer need align with your organization’s strategy? Just because there is a need doesn’t mean that every organization should fill it. Does it align with the core strengths and strategy of an organization question? this is the same question all product people should be asking with whatever product-related effort.
What customer behavior do you seek from your data viz? Once you have identified a need and that your organization’s strategy aligns with that need, then the question is, what is the behavior you seek as a result of the data viz. Generally, this means either wanting a customer to do something or stop doing something, but it could also be having the customer feel something.
Does the data viz fit into the overall product experience? The overall product experience is more important than ever as customers become more sophisticated. Understanding how analytics and data viz fit within that product experience is important for maximum benefit to be received.
Is the data viz good? You can’t forget to do good data viz. No, 3D pie charts with 20 slices. Yeah, the best practices of data viz still come into play. Don’t do the hard work upfront and drop these at the end.
Are the assumptions and behaviors still as anticipated? It is critical that assumptions and behaviors be tested and retested on an ongoing basis to ensure the data viz is functioning as planned.

I call this the Data Viz in Product framework, and as a person who spent many years in product and data viz, this framework has helped me, and I hope it does so for you. Feel free to use it how you like, and I certainly love to hear your stories if you do on what worked and what didn’t.

Now let’s get into data viz and product examples. We will first go through a series of examples and do a mini-dissection of what we think was done well and what opportunities exist. Sometimes, we might have questions. Our belief is that all data viz is part art and part science, so there are no right answers.

Audible App: Starting out, I want to say that I am a huge fan of Audible. It has allowed me to consume more information in a way that is both convenient and also sometimes more emotionally resonant. In fact, I have used Audible for over 10 years consistently. But, sorry Audible, but your data viz sucks, in my opinion!

They have several uses of data viz in their app and include a badge page which, while aesthetically pleasing, seems less useful and more ego wall, but I have heard from others that this is indeed a feature they like. There is also a basic listening time analytics page that provides a rudimentary understanding of listening and provides it on a daily, weekly, monthly, and total basis. There is also a “Listening Level” page, which again seems more like a badge-like page. It appears from this page that simply the number of hours I listen in total is equal to a “Listening Level” and uses terms like Newbie, Novice, Pro, Scholar, and Novice. But, the data viz I am going to dissect here is the “Audible Titles” page, and the graphic from my phone is below.

Source: Audible app screenshot

What is the purpose of this data viz? Honestly, I don’t know, and here are some questions I asked myself:

Is it meant for me to feel better about myself? Maybe it does a little, but it also makes me feel bad at the same time because I know a number of these books I started and didn’t finish, or there are some I didn’t even start.

Is it meant for me to buy more books? I don’t think so because my number of books seems pretty high.

Is it meant for me to enjoy the app more? I don’t think so because I feel there is not much to do in this visualization.

What does Audible do well?

Audible uses an acceptable data viz given it is booked over time. I might have chosen to go with a line graph or something else instead of an area chart, but an area chart can work.

What could Audible do better?

Audible could better understand its users and, if appropriate, use data viz to help its users understand their problems. An individual data viz needs to start out with a desired audience(s) and a desired purpose(s). I don't clearly understand either in this data viz.
Audible could better integrate its data viz into the overall product experience. This does not just make it look visually aligned (which it does somewhat) but also aligns it with the user's overall experience. I feel like this data viz was incorporated as basically a check-the-box around showing books purchased over time and thought, let’s incorporate it into a data viz.
Audible could make better data viz by both enhancing the color and size of the labels on the axis and also reducing visual fatigue of the orange in the area chart.

I think Audible has failed on both (1), potentially (2), (3), potentially (5), and likely (6) with respect to the Data Viz in the Product Framework outlined above. Maybe instead, Audible needs to think about its different user personas and break down how data viz could benefit them and align with Audible’s brand. Audible has loyal customers who are curious people like myself. One thing I might be interested in understanding is when I listen, how much I listen compared to others, etc.

Ring App: Ring is a doorbell with a video camera, speaker, and microphone. It is a product that falls into the Internet of Things (or IoT) space, which I find fascinating and have been involved with in the Twin Cities over the past 5 plus years. Ring is a classic IoT product in that it takes in a ton of data while seeking to provide a service to end users. Ring allows me, as a homeowner, to understand when someone is at my door and can even communicate with that person no matter where I am. It gives me peace of mind when I travel, but it also allows my technical side to be fascinated with the opportunities.

Image Source: Ring app screenshot

Now let’s get to the data viz in the app. Ok, yes, this is my neighborhood, and really, it is a great neighborhood in the Twin Cities with a diversity of cultures, residential and commercial, and a great location. Many others and I look at Richfield as a safe, close suburb of Minneapolis.

From the Ring data viz you might be highly concerned about living in this area. It reports all the crime that is reported and does not delineate between violent and non-violent crime easily; it does not delineate between residential and commercial crime easily, it does not delineate anything with respect to time of day or relative population count and other similar metro areas. Basically, it gives me a map with points laid on it to make a judgment for myself.

I wonder how many people use this data viz for Ring. I also wonder what type of people use this data viz for Ring. I further wonder how many people would value Ring higher if they had a more meaningful level of data viz.

What do you think Ring’s objectives are in showing me this data viz in this format? Maybe it will scare me into buying a more premium ongoing monitoring service. Maybe it is a check-the-box effort, and only a limited amount of effort is put into this.

What does Ring do well?

Ring provides fairly comprehensive data related to the area where I live and not only in the app but also sends out a notification that the report is ready.
Ring arguably leaned in most with its audience that likely uses its product out of fear and protection and guessing a desire by its customers to know all crime around them.

What could Ring do better?

Ring could help users be able delineate data for users to understand their area both as a novice and a nerd. Things like: a) violent vs. nonviolent crime; b) commercial vs. residential; c) give me context on time of day; d) allow me to incorporate traffic patterns or allow Ring to understand my traffic patterns and give recommendations based on these; and e) context to whether the crime is a lot per person for type of area, increasing/decreasing, and other information that may help me make better decisions.
Ring could also help not just lean into fear, though, and maybe provide other delightful information about the neighborhood. For example, my Pocket Casts podcast catcher app provides some humorous items, like 11.47 Trillion emails that were sent during the time you listened to podcasts. Things like this can provide humor and delight in an app and help users that might help bring users back from undesired negative tendencies like unwarranted fear and potential biases.

Fitbit App: I love my Fitbit. I actually used to love my Fitbit more and what I am showing was the prior UI prior to a recent update that I actually think is a worse use of data viz in their product. No matter what, though, Fitbit is something that I have used for years, and it tracks things like your steps, your sleep, and other habits related to your health and wellness. I can easily say it is the data viz in an app that I use more than any other where data viz is not the primary purpose of the app.

Source: Fitbit app screenshot

What does Fitbit do well?

Fitbit understands what users like me seek to use it for by quickly making metrics easily understandable and how I am tracking to my goals and does this on a daily basis. It even provides some easy-to-use insights where I can provide feedback on if I liked it or not.
Fitbit understands that some of its users are novices who will use it to understand if they hit their step goal or other goals, but it also understands some of its users are nerd users who want to be able to dive into details about the information. In addition to the detailed screens Fitbit provides, it provides the ability to download the data into .csv even. Great for a nerd like me, and yes, I have done this.
Fitbit understands that sometimes you need to use multiple encoding to help ensure the user easily understands what you are conveying. When saying multiple encoding, it means relaying the same information in multiple ways. For example, you may encode information with color, shape, size, etc. Fitbit goes in and uses color and shape in how it fills items to let me know when I have hit goals, for example.
One newer feature not shown above is Fitbit has created a sleep number itself on how well you slept. I am still in the process of trying to understand it, but it seems like 0 is the worst and 100 is the best, so it provides you with a sleep number in addition to hours. This sort of allows a novice to go a little deeper into understanding sleep without breaking down the different phases of sleep and the number of total hours.

What could Fitbit do better?

Fitbit UI, prior to the recent update, had the top cards be able to flip days without flipping corresponding data below, which meant a misalignment of data showing different items for different dates. This has been resolved, though, in a recent update, but it is a good reminder for us to have interactivity be well thought out.
Fitbit could do a better job leveraging notifications in tandem with the data viz. This is less a critique of the data viz and more on how notifications of information are often more meaningful than the data viz itself because the notifications, in theory, should be more meaningful nudges.

These are just some of the ways Fitbit has gone in and designed data viz in its app in a thoughtful way that provides me both delight and value.

Mint App: Mint is an app where you can bring your financial information together, and it helps you understand your financial information and better plan and make decisions. I am not a Mint user myself, but I have talked to dozens of people who are, so I am speaking of this from a less biased perspective.

When looking at the screenshots below, you can see there is a lot of good information, with the screen on the left giving you an idea of where you are tracking the total budget for the month, along with how you are tracking individual categories and subcategories. They use color to help you understand how you are tracking. Then, you jump to the middle screen, and it is a donut chart where you have your monthly spending broken down. Lots of colors, not completely labeled, and no delineation of categories that are variable versus fixed costs. The screen on the far left is a simplified view of how your credit score is and an indicator of where it falls on a credit score scale.

Source: https://newswire.newscoop.pro/en/2/1/578/Students-Intrigued-by-Stock-Market-and-Finances.htm

What does Mint do well?

Mint does a lot of great tracking of financial data and helping display it in generally good data viz (ignoring bad donut chart in middle).
Mint uses generally intuitive colors, although it could better use hues to ensure there are no issues for those with color blindness.
Mint does a good job of separating out data viz in different screens and not forcing those together because different cognitive loads can be placed on each.

What could Mint do better?

Mint could help differentiate between fixed and variable expenses because I generally cannot make fixed costs changes easily but variable costs I can.
Mint could help me better assess how I do against others in my peer group and relay this information to me. Not to make me feel overly good or bad but to understand where I compare and maybe also against a type of person I want to compare against.
Mint could drop the donut chart in the middle and put it in a more useful form similar to the screen on the right or put it in a waterfall chart.

Overall, I think Mint does a solid job with data viz in its product and incorporating it into its overall experience, helping around customer problems and helping modify behavior, but at the same time, I think it can improve.

Sleep Number: Sleep Number is a high-end smart mattress company that does some really cool things with its product, but it also does cool things in how it uses data viz. Sleep Number is probably best known for what its name indicates, i.e., its sleep number. It simplifies the sleeping experience to a sleep number and believes it is different for different people, and the beds allow adjustment for each side of the bed.

Source: https://www.sleepnumber.com/

Sleep Number's use of data viz in its product is kind of unique in that what I am showing is not even its use in the product itself, but instead, it is a data viz that is part of the Sleep Number selling experience. You go into a Sleep Number store, and as part of it, you can look at how you sleep and how adjusting the sleep number could relieve the pressure points you have while sleeping. It is a simple data viz where it shows the two people lying down and where pressure is and uses color to differentiate pressure. There is also a Sleep number that is displayed. There is also a before and after view of things.

What does Sleep Number do well?

Sleep Number understands that its customers are more sophisticated and want to feel they are buying a premium mattress that helps them sleep better. Harnessing data viz, in this case, is to help them do just that.
Sleep Number understands that in-store salespeople have a limited time to close a sale and are often of mixed sales expertise, and leveraging this data viz helps empower its salespeople on both fronts.
Sleep Number adds a level of credibility and authenticity to its product, and seeing is believing. Using data viz in this capacity helps Sleep Number better support its product value proposition.

What could Sleep Number do better?

In my opinion, there are no clear opportunities with respect to this data viz. Certainly allow the salesperson to print or email this data viz to the person, especially in the case where the sale didn't happen, so maybe potential follow-up.
I'm not sure if this data viz can be used in the hope, but if so, having that ability would also be great so people could adjust this with their app.

Data Pine Google Analytics Dashboard

Marketing dashboards are common, and this one is Data Pine’s Google Analytics Dashboard, which helps users understand and monitor Google Analytics data for one or more sites.

There are a ton of examples of good and bad dashboards out there, and this is a good example.

What does Data Pine do well?

Right across the top are cards that are labeled but also have relevant icons and colors. Knowing that items at the top of a page get the most eye attention makes a lot of sense, assuming these cards are indeed the most important items for consumers.
Cards are aligned together on the dashboard to tell the "Google Analytics story," which is how the metrics measure performance.
Good use of data viz practices in carrying out visuals.

Source: https://www.datapine.com/google-analytics-connector

What could Data Pine do better?

The size of text is fairly small in some areas, so enhancing the size of text and, if it doesn't fit, for example, then allowing a shortened view of text that is larger would be beneficial.
There seems to be some confusion with the labels on the cards at the top of the dashboard and in the charts below. They are using the same label but seem to show different data.

Hopefully, these examples were helpful for you. Remember, data visualization is part art and part science, so talented people may disagree. The above are just opinions and not meant to endorse products or capabilities.

If you are a product person, then we hope in the future, you will pay closer attention to how information is being relayed in or with your product. Not everything needs to be a chart or a dashboard to relay data. However, thinking about leveraging concepts of gamification, behavioral science, and user experience as part of relaying this information is essential.

Good luck in incorporating data viz in your product, and I hope you leverage the six-step Data Viz in Product framework above so you more consistently deliver valuable information in your products that align with your users and your desired experience.

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

Ep 34 - Lori Silverman - What it take to drive analytics adoption

Beyond the Data — Wed, 16 Oct 2019 13:32:54 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“What is the goal of bringing data into organizations? People make tens of thousands of decisions everyday... data can create more intelligent groups that improve those decisions.”

— Lori Silverman, CEO & Professor

Combining multiple disciplines can so often be a recipe for success. Some of the best inventions, ideas, and movements were started because someone from a different background came in to an industry, saw things through a different lens, and approached a problem from a completely different angle.

For Dave, that “combination” has been through Chemistry, Law School, Product Management, and now Analytics.

For Matt, that “combination” has been through Computer Science, Industrial-Organizational Psychology, and now Analytics.

For Lori Silverman, she brings a wealth of experience from her psychology background, but has also worked on many “up-and-coming” movements in the quality, organizational change management, and storytelling spaces.

In this episode, Lori shares incredible insights from 30+ years of helping organizations think differently, improve results, and “shift” the way they do business.

Lori is passionate about driving change and about helping her customers succeed in building data informed organizations. She’s also a consummate researcher, so she’s always looking at ways to combine research into “real-world”. We talk a little bit about the research behind an insight (literally called an “ahas"!”). We also talk about the research behind System 1 and System 2 thinking, developed by the amazing Dr Daniel Kahneman and Dr Amos Tversky, summarized in their book Thinking Fast and Slow.

If there’s one thing Lori does, it’s to inspire people towards becoming a champion for data. She tells us the story from early in her career, in Wisconsin working for a local hospital, where she identified a need, went out of her way to learn from great people on the topic, and then spent 2 years slowly building buy-in from various leaders to change how the organization thinks. This is exactly what data champions do! They influence their peers and get buy in from leaders to help facilitate the change they know is possible!

Check out the whole episode for lots of great tips and inspiring stories from a true thought leader!

More about Lori Silverman

LinkedIn - in/lori-silverman-700963
Website - www.partnersforprogress.com
Twitter - @LLSilverman
Lori’s Books - Business Storytelling for Dummies, Wake Me Up When the Data is Over
Heroes - Cassie Kozyrkov - Chief Decision Scientist at Google

Learn more

Permalink

What are your Data Principles?

Beyond the Data — Thu, 10 Oct 2019 11:00:00 +0000

The first thoughts you might have are: what are Data Principles, or maybe are Data Principles really a thing? Think of Data Principles as your manifesto or by-laws for how your organization uses data. They can be an important right in setting the tone of your organization in how data is used from the C-level throughout. Data Principles could be about how you go about collecting data. How do you go about sharing data within the organization? What is your view of data in the decision-making process? What are your views on Data Governance? What is transparent, and what is hidden from your consumers, partners, employees, etc? Certainly, Data Principles can include a lot.

Let’s take a step back, though, and provide a little more context on Data Principles, we shamelessly adapted this concept in part from Ray Dalio’s Principles book, which is a great read. Ray Dalio applied Principles that he used in his personal and professional life. While I recommend you read the book, I think the concept of putting together your Data Principles for three reasons:

The process of developing principles helps engage in a meaningful conversation and ensure there is buy-in.
Provides an immediate reference point so that everyone in the organization can understand how data should and should not be used within the organization.
Better provides accountability in using data well in an organization and, at the same time, helps democratize the message around the importance of data.

There are certainly other reasons that these Data Principles matter. One important point when putting Data Principles together is that this should be done by a diverse, cross-functional, and cross-seniority level group. It can’t just be leaders. It can’t just be data people. A broad group of input and agreement on Data Principles is needed. Further, the result of this effort needs to be even more broadly communicated so that everyone understands the result.

The process of coming up with Data Principles is just as important as the Principles themselves. Don’t cut the process short; use it as an alignment and communication effort around how your organization cares about data and will use it strategically and ethically. Having an outside facilitator helping provide the Data Principle development sessions is often beneficial so that it is not too driven by one area or another of the organization. It is important that all areas feel and actually do have input in the Data Principles coming together.

You might ask, "What should our Data Principles be?” I cannot answer blanketly, but instead, each organization’s culture, maturity level, risk tolerance, and other factors come into play. What is important is that the conversation happens and the Data Principles result. One important point in coming up with Data Principles is not lying to yourself as an organization. Data Principles should align with the culture or, if different than the culture, then align with the direction that the organization is broadly going to put effort into shifting culture.

Data Principles are only as good as the effort you put into creating, communicating, and reiterating. If you have your Data Principles defined and want to share them, then I would love to see them.

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

How Data is Helping Leaf Peepers Plan the Perfect Fall Trip

Beyond the Data — Tue, 08 Oct 2019 10:00:00 +0000

Happy Fall! Fall is my favorite season and living in the Northeastern region of the Continental US I look forward to chillier days and leaves changing. I enjoy seeing what Albert Camus describes as “Second Spring” where the leaves are a vibrant red, yellow and orange! The challenge I have every season is finding that right time to drive up to that scenic overlook to capture a picture perfect landscape of colorful leaves with just the right amount of blue skies. It looks like a data scientist, Wes Melton might have solved our problems by determining the precise future date that the leaves will peak in each area of the Continental US!

Wes Melton along with his co-founder David Angotti have created the Fall Foliage Prediction Map. According to Melton, this is one of the only fall leaf tools that provides accurate predictions for the entire continental US. Looking at the snapshot above, a user can take the slider across the 12 week period to see when and where fall foliage will peak in a given region. Looking at the week of October 5th, we can can see patches of the Northern US hitting peak season like Maine, Vermont and New Hampshire. Another great area that might not come to mind are the aspen trees in Colorado.

The story behind the creation of this map is interesting. At the time that they were starting their Smoky Mountains Cabin rentals six years ago, they were getting questions on the best times to experience the fall colors. They started conferring with meteorologists and the predictions were accurate. The advice was well received and they have been updating the tool ever since!

“Autumn is a second spring where every leaf is a flower”

— Albert Camus

To create the map, a complex algorithm was developed that carefully analyzes several million data points and outputs approximately 50,000 predictive data pieces. This data then enables the program to forecast county-by-county the precise moment when “fall peak” will occur. As time goes on and the algorithm is fed more data, it will only become more and more intelligent.

Some of the data points processed by the prediction algorithm include National Oceanic Atmospheric Administration (NOAA) historical temperatures, precipitation, forecast temperatures, and forecast precipitation; historical leaf peak trends; and peak observation trends.

I am excited about this story and this tool because it was birthed from customers looking to rent cabins and Melton and Agnotti went above and beyond to accommodate them. They didn’t stop at the Smoky Mountain Region but went on to answer this question for the whole continental US. They were also able to compile data points from several sources to piece together this aesthetically simple visualization.

Check out the map and learn a little bit more on how and why the leaves change color. I would love to hear about your favorite leaf peeping location!

About the Author

Allen Hillery

ADJUNCT PROFESSOR AT COLUMBIA UNIVERSITY,
WRITER AND EDITOR AT NIGHTINGALE, A MEDIUM.COM PUBLICATION

You can sample his work here: Three Reason Why Storytelling is Important in Business

Follow Allen:

Website: https://medium.com/@alglobehopper
LinkedIn: Allen Hillery
Twitter: https://twitter.com/AlDatavizguy
Publication: https://medium.com/nightingale

Ep 33 - Matt Anderson - The Link Between Librarians, Product, and Data

Beyond the Data — Tue, 01 Oct 2019 11:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“A lot of product managers are looking at their sales numbers, but they’re not thinking broadly about how the data can provide a wider lense.”

— Matt Anderson, Product Manager

Using data is critical for every facet of the business. But none is more powerful and readily usable than in product management.

Product owners, product managers, product analysts, you name it. Companies who have taken the plunge into digital transformation and agile frameworks need great product people. And those great product people MUST rely on data to do their jobs.

Matt Anderson didn’t start his career in the product space… he started as a librarian! But he found his niche in product management and has been using data to help understand his customers, his product, and his vision to drive profitability and sales for his company.

In this episode, we talk about what data can do for business folks… both how to use it, and how NOT to use it.

More importantly, Matt talks about his unique approach to collecting data that feeds the questions he’s trying to answer. This is different than the typical approach of “use whatever data you have”. Instead, he’s thinking strategically about what data he NEEDS, then goes and gets that data from his customers. He’s also passionate about QUALITATIVE data, not just quantitative. The user’s own stories are what provide the context that helps shape where the product can go.

My favorite story from our discussion was when Matt talked about using data to NOT make a decision. See, often times we think about data informing a decision… to take an action in some way. Matt found that his data collection efforts actually helped him steer clear of a decision that may have been costly.

Make sure to follow Matt on LinkedIn and Twitter, as he’s regularly writing about relevant data + product topics!

More about Matt Anderson

LinkedIn - in/matt-anderson-87988823

Twitter - @MattAndersonUT

Website - www.mattanderson.org

Heros - The folks from the New York Times Graphics

Favorite Book - The Lovely Bones by Alice Sebol

Great Storytellers - John Cutler, Melissa Perri, Theresa Torres

Would you like us to email

when new episodes air?

Permalink

Data Literacy is Not One Size Fits All

Beyond the Data — Tue, 24 Sep 2019 14:15:40 +0000

“You don’t have to be a data scientist to be data literate.”

In fact less technical colleagues are welcome to help narrow the data literacy gap! As I mentioned in my last post, there are 2.5 quintillion bytes of data created daily. I love Chartio’s view of the current BI landscape - The world has gotten really good at collecting data, now the largest bottleneck is our ability to understand the data and make informed decisions based on it.”

“You don’t have to be a data scientist to be data literate”

There’s a lot of data to process and most companies have been hiring the most technically oriented people they can find to build armies of analytics and data science teams to analyze data. The one thing they have ignored is the data professionals’ ability or desire to communicate with a general audience.

“The world has gotten really good at collecting data, now the largest bottleneck is our ability to understand the data and make informed decisions based on it.” — Chartio’s view of the BI landscap

I’ve worked on several analytics teams and while I choose to champion the capabilities of data, I’ve seen my peers struggle working with business teams or fall short in explaining their analysis. I’ve also found the expectations placed on analytics teams to be unrealistic at times. Analysts are expected to wrangle data, analyze it in the context of knowing the business and its strategy, make charts and present them to business stakeholders with short turnaround times. Wash, rinse and repeat.

The bump in the data road lies right at the last mile - when it comes time to explain the analysis to decision makers. In a question on Kaggle’s 2017 survey of data scientists, to which more than 7,000 people responded, four of the top seven “barriers faced at work” were related to last-mile issues, not technical ones. Here they are in the word cloud below.

I’ve experienced all of the above; I can assure you all that it’s not fun. What I have come to realize is that data is not just for the uber technical. The opportunities in data can be harnessed by many with liberal arts backgrounds. Before you gasp in disbelief, hear me out. The identified bottleneck has been the ability to understand data and make informed decisions. Combined with the four barriers that have been cited above we need individuals who can narrow the data literacy gap by:

Framing questions correctly
Bringing together cross functional teams to work effectively in analyzing data
Communicating results to decision makers and the public

Analytics teams need all the help we can get at the last mile! While you may have believed that without knowing what an R package is there is no way you can contribute to a data project, you couldn’t be more wrong. When it comes to analyzing and presenting data, critical thinking is crucial. If you’re on the business side of the organization, you are closer to the key performance indicators that the company is striving to obtain. You could potentially be a project manager with a proven track record of meeting deadlines. These are all skills that are much needed to drive data analytics pass the finish line!

One trend that has been growing in data driven organizations is hiring of liberal arts talent. These individuals possess a lot of the key skills needed for analysis - critical thinking and context setting. I like what William Cronon writes in his article, “Only Connect”. He defines a liberally educated person as someone who can:

Data folks, I’m not taking anything away! The data presentation piece of the puzzle needs to catch up to all the advancements we’ve made in ingesting and processing data. These additional talents will complement our teams and the symbiotic relationship will advance our cause. In Scott Berinato’s article, “Data Science and the Art of Persuasion”, he points out that one of the steps to building a better data science operation is to define talent not team members. The core set of talent that Berinato describes is qualities I’ve seen in past teams. They include:

As Berinato pointed out in the article, there’s a difference between talent and team members. A team member can possess a few of the talents listed above. I know that I’ve strived to be an ambassador in my organizations and bridge marketing and analytics folks to move projects along. I’m also very passionate about presenting data insights as a story.

About the Author

Allen Hillery

Adjunct Professor at Columbia University,
Writer and Editor at Nightingale, a Medium.com Publication

You can sample his work here: Three Reason Why Storytelling is Important in Business

Follow Allen:

Website: https://medium.com/@alglobehopper
LinkedIn: Allen Hillery
Twitter: https://twitter.com/AlDatavizguy
Publication: https://medium.com/nightingale

Ep 32 - Catherine D'Ignazio - Getting the Data Basics Right

Beyond the Data — Wed, 18 Sep 2019 10:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“As an educator, I am always working with people who aren’t naturally “numbers people”. I believe that you don’t need to be a data scientist to effectively work with data.”

— Catherine D'Ignazio, Data Educator

What does feminism and data science have in common? Well if you talk to Catherine D’Ignazio, quite a lot actually!

Catherine was in Minneapolis for the Eyeo Festival over the summer and Dave sat down to learn more about her presentation, some of the work she does as an educator, and about some of her side projects like the “breast pump hackathon” and the Data Literacy tool, “Data Basic”

Obviously we had to dive into the hackathon a bit more to understand exactly what that was, and how it came to be (it’s actually a really cool cause!)

But Catherine’s work in data literacy was what got us really excited.

Catherine co-created DataBasic as a suite of easy-to-use web tools for beginners that introduce concepts of working with data. These simple tools make it easy to work with data in fun ways, so you can learn how to find great stories to tell.

Dave also talked to Catherine about data journalism, something that Catherine spends a lot of time in. They talk about the mission of journalists to provide unbiased information, and how data can be such a critical piece of doing that well in the future.

More about Catherine D'Ignazio

LinkedIn - in/catherine-d-ignazio-61a57ab1
Twitter - @kanarinka
Website - www.kanarinka.com
Passion Project - makethebreastpumpnotsuck2018.com
Data Basic (Data Literacy) - databasic.io

Permalink

Why We Should Be Excited About Data Literacy

Beyond the Data — Tue, 10 Sep 2019 10:00:00 +0000

Hello! My name is Allen Hillery and I’m happy to be teaming up with Matt and Dave to get you excited about Data Literacy. I’m a data champion who has worked with business and data teams throughout my career playing the role of ambassador and coaching them on how to better leverage data. I’ve had the opportunity to work in companies with varied data maturities ranging from reactive to more thoughtful on executing results. Like most of you, I aspire to work in a truly data informed organization where everyone is literate to understand the context of their data they’re analyzing and the value it brings internally and externally.

So my question to you is - How comfortable are you with data? Does the thought of getting your hands dirty with data excite you or make you want to cringe? According to Forbes, there are 2.5 quintillion bytes of data created daily. If you think about it, data is a major part of our lives. Each one of us, generates data as we move from google searches to shopping with a club card at the supermarket, not to mention data created by Internet of things. In the office, are you the go to dashboard expert or maybe you’re resident data whisperer who massages insights out of your analytics teams?

Being data literate means you have the ability to read, understand, create and communicate data as information. We are on the precipice of an exciting time, as we have superfluous data available to analyze. This data can present information that provides better customer experiences and enables your team to identify which segment would be best served by your products. While the amount of data being created can sound daunting, the evolution of the tools and infrastructure to help us navigate this landscape is intriguing!

“People aren’t going to go to BI, BI has to go to to the people.”

— Nick Caldwell

Tech executive, Nick Caldwell said, “People aren’t going to go to BI, BI has to go to to the people. This is already happening in a big way.” The staggering amount of data that has been made available to us has hit a tipping point where data analysts have to enable non technical business partners to develop insights on their own. This trend has caused a shift towards more intuitive self-serve tools. At the same time, the proliferation of opportunities to learn query language are seemingly ubiquitous.

In addition to trends pivoting our work cultures to being more data informed, the growth and learning opportunities that will come from leveraging both data and data literacy have me really psyched! Companies are beginning to realize the importance of investing in their employees’ data literacy. AirBnB is a shining example of investing in data literacy through the creation of their data university. This effort was made with the belief that every employee should be empowered to make data informed decisions. It took roughly two years to launch but one of the amazing results is a reported 50% increase in active use of their internal data platforms. Another benefit is that it frees up data teams to concentrate on more complex tasks.

Sharing success stories, like AirBnB illustrate the importance of empowering employees and customers with data. Think of all the apps and services you use right now. You’re leveraging data when you are booking that next AirBnB, searching Yelp for food recommendations and hailing your lyft to get around. BI is coming for you and you’re more acquainted with data than you realize. So maybe you’re the resident data wrangler on your business team who realizes that data is not as aloof or mysterious as you once thought? Maybe your knowledge of the business combined with your new found data sleuthing skills has put you on a direct path to being a data champion lobbying for more training? Then you’re at the right place! We’re here to reassure you that you don’t have to be a data scientist to be data literate! You just have to be open to getting your hands a little dirty with understanding how to leverage data!

About the Author

Allen Hillery

Adjunct Professor at Columbia University,
Writer and Editor at Nightingale, a Medium.com Publication

Follow Allen:

Website: https://medium.com/@alglobehopper
LinkedIn: Allen Hillery
Twitter: https://twitter.com/AlDatavizguy
Publication: https://medium.com/nightingale

Learn more

Ep 31 - Tricia Duncan - Implementing Data Viz in Organizations

Beyond the Data — Thu, 05 Sep 2019 10:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“The most data-informed organizations I’ve seen are the ones that have a plan, and that integrate data into their day-to-day, instead of using it as an afterthought”

— Tricia Duncan, Data Luminary

As analysts and “data people” we often see all the amazing things that are possible with data, data science and data visualization. We research new tools, new technologies, and new approaches.

But we often work for organizations who are “stuck in their ways”, content with that excel table instead of a sankey diagram. This can be frustrating when you SEE the possibilities, but you can’t convince anyone to move in a better direction.

So what do you do? Is it you? Is it your organization? Is it the leadership?

In this episode of Data Able, we talk with Tricia Duncan who has been consulting on Tableau, Data Visualization, and new approaches for over 6 years. She’s worked with small, mid-size, and fortune 500s all over the midwest to help them implement data visualization best practices and truly “modernize” their approaches to analytics.

As someone who has seen all sizes and kinds of organizations, we were interested to see what kinds of roadblocks existed. Is everyone as averse to modern BI and visualization approaches, or is it just a select few?

What Tricia has seen, leads us to beleive that this is a common problem, not limited to any single team, industry, or size company.

One of her stories revolves around a Chief Marketing Officer who wanted to see some new marketing numbers. Tricia saw the opportunity, built an amazing dashboard, and was met with confusion by the CMO when delivering it back.

While her dashboard was likely “better” than what the CMO wanted, it didn’t match the intended ask. The valuable lesson Tricia (and we) learned was that its better to deliver on the ask, and “slow-feed” people a more visual approach. Give them a little bit more each time they ask for something. Getting them from 1 to 2 on the maturity scale is far easier than trying to get them from 1 to 9.

Check out the whole episode for more great tips on how to help your organization improve their analytics maturity!

More about Tricia Duncan

LinkedIn - in/triciaduncan1

Links from the episode

Data Hero - Nick Pedersen - Planning Director @ State of MN
Favorite Book - The Model Thinker
Favorite Storyteller - Octavia Butler
Favorite Podcast - Partially Derivative

Learn more

Permalink

Do your metrics have a positive ROI?

Beyond the Data — Tue, 03 Sep 2019 11:00:00 +0000

Metrics are important, but not everything needs to be measured in a data-driven world. Adding a metric without understanding the upstream and downstream consequences is easy. That is why it is important that your metrics have a positive return on investment (ROI).

ROI is a simple calculation where the benefits are divided by the cost, and if the benefits are greater than the cost, you have a positive ROI. This might sound obvious, but most of the time, metrics are implemented in silos, and upstream and downstream impacts are not understood and accounted for.

So, what are the potential benefits and costs of metrics:

Potential benefits:

Enhanced clarity by leadership resulting in enhanced decision-making outcomes
Changed behavior that results in more revenue-generating activities or reduced cost-generating activities
The benefit of not having to maintain other metrics and related costs if a metric is replacing one or more metrics
Reduced friction between areas with metrics that align teams or departments with organizational objectives

Potential costs:

Time spent in calculating metrics by individuals
Storage and processing costs in creating metrics
Time spent rolling out new metrics
Time spent communicating and reviewing metrics
Changed behavior that results in less revenue-generating activities or increased cost-generating activities
Enhanced friction between areas with metrics that misalign teams or departments

Each metric that is implemented should have a clearly positive ROI. The purpose of calculating ROI is not just to come up with a precise measure though. We think calculating ROI related to new metrics is most valuable because it provides a process for deliberate thought around implementing new metrics and maintaining old metrics.

Just as important as determining the ROI on new metrics, you should ensure the ROI on existing metrics is still positive. While metrics are not something that should be changing all the time, they should be reviewed and updated deliberately.

Hopefully, this post helps you think more about your metrics and their ROI. Much of being a data-informed organization is simply the critical data thinking that goes about aligning to the desired organizational mission.

This is the first in a series of posts we have planned around metrics. I think metrics should be front and center in an organization that values data because metrics are something that people are already familiar with and use regularly. More importantly, I think good, communicated metrics can unite and accelerate an organization.

Looking to implement metrics & KPIs in your organization and need help? Make sure to contact us and see if I can help.

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

Ep 30 - Nadieh Bremer - Anatomy of a Great Data Visualization

Beyond the Data — Tue, 13 Aug 2019 11:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“Want to be a better data visualizer? Make lots of projects. Look for other people’s work and try to iterate on it. Pick something you’re passionate about and start making something.”

— Nadieh Bremer, Data Visualization Freelancer

We hear a lot about people transitioning into a data science role.

But how many people have you heard who are transitioning OUT of data science and into something more artistic.

Meet Nadieh Bremer, an ex-Deloitte data scientist with a background in astronomy and predictive algorithms.

Nadieh is a leader in the data visualization space, but she didn’t always start there. After years of churning out “just another predictive model” she was in search of something that fueled her more creative side. And she found data visualization! She didn’t realize just how powerful and needed these skill-sets really were.

Nadieh now does data visualization work full time through her company, Visual Cinnamon. She has won data visualization awards for her work in such publications as Scientific American, The Guardian, World Bank and Google News Lab. We also highly recommend checking out her visualization on Lord of the Rings!

We asked Nadieh to walk us through her process for creating the Lord of the Rings project. Surprisingly, there was much more to data visualization then just creating a pretty chart! Much of the data that she needed to answer her question wasn’t available in a format that was useful.

Hear her describe the effort from start to finish, and learn how to create awesome visuals that both captivate and inform!

More about Nadieh Bremer

LinkedIn - in/nbremer
Twitter - @nadiehbremer
Nadieh’s Website - Visual Cinnamon

Links from the episode

Dataviz - Lord of the Rings Project
Project - Data Sketches: A Year of Exotic Visualizations

Permalink

Ep 29 - Ben Schein - Organizations Need More Data Curiosity

Beyond the Data — Tue, 30 Jul 2019 10:30:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“I don’t want to ever build anything that’s completely done. I want to leave that last mile unfinished because it enables lots of people to answer lots of business questions.”

— Ben Schein, Vice President at Domo

What if you had all the best data lakes, ETLs data pipelines, and BI tools in your organization?

What if you had an amazing team of technical data experts capable of writing python, R, SQL, and proficient at data visualization and data storytelling?

That would be great, right? You’d have a well-run data organization! Except… maybe you wouldn’t.

On this week’s episode, we talk with Ben Schein, VP of Data Curiosity from Domo. What an awesome title!

Ben is on a mission to show that simply having skills and tools is NOT enough to a data-driven organization in today’s world. As a former data leader at Target Corporation, he saw that the quality of the business team’s questions really mattered. If they were curious, asked lots of questions, and sought out insights, those teams would be most successful in implementing a data driven culture.

Of course, data curiosity is a two-edged sword. If you can’t deliver on the questions, that’s not good either. Ben’s solution was brilliant… build your data products end-to-ALMOST-end, leaving the last mile available for the business teams to scale their questions (and their answers).

Check out the whole episode for some amazing tips and stories about empowering teams with data, and developing a “data curious” workforce!

More about Ben Schein

LinkedIn - in/ben-schein
Twitter - @benfrominn
Ben’s company - Domo

Links from the episode

Book Recommendation - The Idea Factory
Mentor - Paritosh Desai, Chief Data and analytics Officer at Target
Mentor - David Hussman, Founder of DevJam
Mentor - Jason Goldberger, CEO at BlueNile

Learn more

Permalink

What kind of games do you play?

Beyond the Data — Tue, 23 Jul 2019 11:00:00 +0000

You might answer this question with chess, basketball, Call of Duty, Overwatch, tennis, poker, or Dungeons & Dragons. But now, if I ask you, “What type of game is your current change or transformation effort at your organization?” What is your answer? Most people I have asked that question are not sure how to respond. We look at games as something we do for fun and leisure and sometimes will incorporate into a learning activity, but we generally don’t consider games a serious activity.

There is the concept of finite and infinite games that James Carse detailed in his book “Finite and Infinite Games: A Vision of Life as Play and Possibility.” Carse’s main concept is that there are both a) finite games and b) infinite games. Finite games have fixed boundaries and rules, and there is someone who wins, which results in the game ending. On the other hand, infinite games have no fixed boundaries or rules, and the purpose is to continue the game indefinitely with no ending.

One type of game is not better than the other. They both serve a purpose. But, it does matter what type of game is intended and played by those playing.

Friction can arise if you play a finite game in an infinite-game way or vice versa. There are also “advantages” and “disadvantages” you will have if you are playing a finite game and others you are playing with are playing an infinite game.

Regarding my question above, most organizational change or transformation efforts are treated as a finite game. There is a start and an end, and rules are laid out with an objective goal in mind. Accordingly, an organization’s players, namely employees and other stakeholders, look to play a finite game and win.

While treating change efforts as a finite game clearly helps organizations plan and invest, does this approach hinder sustaining change? My proposition is yes. This is because sustaining changes like digital, data literacy, customer-centric, and product-oriented transformations have no finish. They need to be treated as a game that continues and without winners and losers but rather how an organization and its stakeholders benefit from playing the game.

Maybe you buy into this concept, and your next question is, “How do I do it?” Here are some general things to consider as you design your transformation game:

Establish leadership support for an infinite game approach.
Communicate and incentivize people involved in leading the transformation to design and support an infinite game.
Establish and support councils and communities that will help drive transformation and sustain it.
Establish metrics that incentivize an infinite game.

Thinking of your large transformations as an infinite game will help ensure they sustain and have maximum benefit. Good luck with your transformation game, and I hope it is well played.

No matter what happens, keep playing and have fun doing what you do.

- Dave Mathias

Follow me on LinkedIn

Need help?

Contact me

7 steps for getting started with data as a business person

Beyond the Data — Mon, 15 Jul 2019 11:00:00 +0000

Getting started in data can be overwhelming in today’s world. The hype machine of Big Data and Data Science makes it feel like you need to learn 3 different coding languages, 4 BI tools, and have a PhD level skill set in machine learning and statistics.

Case in point. A couple weeks ago, I was in Chicago doing a Data Visualization and Storytelling workshop. After the day-long training, an attendee asked if I could chat. She had been tasked with building some executive dashboards for her small organization and fell in love with data and driving more meaningful analytics. So she started doing some research about moving into the field. But after reviewing several job descriptions, she felt completely lost and pretty discouraged. There was no way she would qualify for those kinds of jobs without going back to school and spending countless dollars and hours.

I assured her there were LOTS of jobs out there for her skill sets, which included Excel, business context, communication, project management, and translating needs into requirements. But, those aren’t obvious to a person just getting started.

If you’re a business person and want to get deeper with using data here are my seven recommendations to get started:

Ignore the hype

There is a real need for great data scientists in this industry. But you’re not going to be one of those. At least not in the next couple years. The good news is that for every 1 data science job, there are 10 business analyst jobs that are just as critical. People who can translate the business. Who can project manage and communicate effectively. Who can drive change management and adoption of data-driven approaches. Those are the jobs for you.

Identify and focus on your unique talents

Focus on the skills you DO have, not the ones you don’t. Perhaps you have an accounting background. Maybe you are great at training people or speaking? Do you like listening and empathizing with your peers and leaders? Find what makes you, uniquely you. I bet your organization will benefit from combining whatever that is with data.

Raise your hand for data-related projects

You DO need some understanding of how data works, and you need to prove it. I once hired a person with virtually no professional experience over a person with 6 years of data 7 analytics experience. Why? Because the no-experience person during a summer internship had raised her hand to do some reporting, found Tableau, tried their free trial and got the whole department to start using it. Organizations want people like that.

Pick a tool

There are a mind-boggling number of data and analytics tools out there. Nobody can learn all of those tools. Pick one and get started. Most of them behave similarly enough that once you learn one really well, you can pick up most of the others well enough. Remember, you don’t need to be everything to everyone. Get good at one thing and build to the rest.

Take a class or workshop

You don’t need to go back to school to be a great analyst. But you can easily pick up some of the basics through cost-effective online or in-person programs. Udemy, LinkedIn Learning, Udacity, and Coursera all offer good self-paced programs for reasonable prices (stay away from the “data science” programs to start). Want a more guided and personal format that’s still cost effective? Check out our programs from Beyond the Data, which are tailored to business professionals like yourself.

Consume lots of blogs & podcasts

There are so many great free resources out there from leaders. Just start reading and listening. You’ll pick up all kinds of useful information on how to become great at using data. Blogs like Storytelling with Data, Flowing Data, and Visualizing Data. Podcasts like Data Skeptic, Data Stories, and Present Beyond Measure. We’re also pretty partial to our own podcast: go Beyond the Data.

Join a community and learn from leaders

Getting connected to other people like yourself is THE best way to jump start your new direction toward a data career. You’ll meet influential leaders in your local area who know the right people and can help you navigate your local market. I guarantee they will be willing to coach and mentor you along the way. That person from Chicago? She started networking and within a week had eight (8) different leaders reaching out to HER about their open jobs!

I can hear it now… “That’s all fine and good, but the job description says I need 3 years of SQL and 5 years of Python!” Yep. HR put those requirements on the job description. One of three things will happen if you apply:

The hiring manager actually needs those specific skills and you won’t get the job. You probably don’t want a job that technical anyways.
The hiring manager is willing to overlook some or all of those technical skills because teaching technical skills is way easier than teaching people/business/soft skills.
The hiring manager just copied and pasted the job description from a different role and doesn’t really care if you have it or not… they were just trying to weed people out.

My point is don’t let yourself be stopped by silly things like “x years of z tool.” You have unique talents that you can bring to the table. Fuse those talents with data, and you’ll be a data rockstar in no time.

Good luck and happy analyzing!

Ep 28 - Rachel Stuve - Fusing People with Data

Beyond the Data — Wed, 10 Jul 2019 11:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“I can teach someone how to code or use visualization tools. But I can’t teach someone to be inquisitive and to solve a problem. The human side of data is so important.”

— Rachel Stuve

The reality of data and analytics today, is that it’s not really about the data or analytics at all. It’s about the human behavior. The choices that executives and leaders make. It’s about augmenting those daily decisions to make them slightly better than if they didn’t have data. Over time, these add up to immense value.

Many analytics teams still focus on the data engineering. The data pipelines. The BI tools to use. Data governance and access to data are undoubtedly important, but it is all for nought if the human on the other side of the dashboard can’t or won’t do something about it.

That’s why I’m so excited to have people like Rachel Stuve in our industry. Rachel believes that data empowers humans: it's what gives us the ability to solve problems and change the world. With data, she believes that unlocking the true power comes from combining the human with that data.

So what does a data informed organization look like?

From what Rachel has found working with organizations large and small, these would be the steps:

They would start with their business goals
They would break down their goals into sub-goals
What are the business questions or challenges that are keeping you from those sub-goals
Link your data to those questions and challenges.
Write the data pipelines, models, code, reports, dashboards and communicate

The key to all of this is that the analytics people are embedded directly in the business, linking data to the business objectives and driving value on the business’ terms.

We couldn’t agree more! Adding value starts with tightly aligned goals. Thanks for coming on the show and sharing your thoughts, Rachel!

More about Rachel

Rachel on LinkedIn - in/rachel-stuve
Women in Technology - https://www.womenintechnology.org/

Links from the episode

Book - Made to Stick: Why some ideas survive and others die by Chip & Dan Heath
Book - Info we Trust: How to inspire the world with data by RJ Andrews
TED Talk - Simon Sinek: How great leaders inspire action
Framework - The Five Whys

Permalink

When Technology and Creativity Collide - EYEO

Beyond the Data — Tue, 18 Jun 2019 15:00:00 +0000

Last week I (Dave) had the opportunity to be back at one of my conferences, the Eyeo Festival. What makes Eyeo Festival great is it brings together people from all over the world around art, creativity, technology, data, and social impact. You get discussions around science, machine learning, social justice, art, data visualization, and more and many of the sessions cross multiple disciplines.

One of the key takeaways for me was learning to create better spaces and communities around us. The cross-discipline nature of an event like this opens up so many avenues to creatively address business and social challenges, and in turn makes it easier for people to harness the power of data.

To give you a sense of just how diverse an event like this can be, here are some examples of the people I met and the conversations I had…

A person came by my table saying he needed our input on visualization. We chatted for a bit, discussing the challenge he was having, without realizing we were talking to famous Bre Pettis, who co-founded MakerBot, a fantastic company that pioneered affordable 3-D printing! Such a cool opportunity to kick around ideas with amazing and talented people!

I had to laugh when we got into a fascinating discussion on better transportation approaches in large cities. Turns out we were talking with employees from both Google and Uber!

This night is just an example of the excitement Eyeo can bring or really any event where you bring diverse, creatives, and technologists together that have a shared mission.

Needless to say there was no shortage of knowledge being shared and absorbed. Eyeo is a reminder for me that I can never stop learning and sharing. The world is always changing and the challenges are always developing where we can play a small part. Find a way to share something with others and always a way to learn something new from others.

Beyond interesting conversations, there were also great sessions to attend. While I don't like to call favorites, one session I found fascinating is Darius Kazemi discussion around the FriendCamp network that he started. FriendCamp is a “local” social network that is intentionally kept very small with only Darius and a set of friends being allowed in.

Darius encourages others to think about creating a similar type of network. This is somewhat of a natural pullback of the not-as-free-as-we-think culture we live in where our data is sold to others and privacy is challenging. He says it is about taking responsibilities of your data and something we should be doing no matter at home or work. If you are not aware of your data, who has access to it and how, and what it is being used for then don't expect anyone else will treat it any better.

While I certainly agree with some of Darius's premises, I also realize not everyone is going to maintain a social network server and rigorously identify and maintain their private social network.

Thank you everyone that came to #Eyeo2019 and the entire team that puts it on because this event is really inspirational. Hope to see you at #Eyeo2020 back in the Twin Cities.

Quick side note: During Eyeo I had the chance to record Data Able podcast episodes with both Nadieh Bremer and Catherine D'Ignazio. Nadieh and Catherine are both phenomenal people putting out great work into the world and sharing information. Make sure to check out their work and subscribe to the go Beyond the Data podcast so you don’t miss their episodes when they drop later this summer.

- Dave Mathias

Follow me on LinkedIn

Learn more

Ep 27 - How Caroline & Sara are building a huge dataviz community in Portugal

Beyond the Data — Tue, 11 Jun 2019 12:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“A lot of people know about data visualization, but they don’t actually understand the possibilities that can be unlocked”

— Sara Mesquita

Something must be in the water in Lisbon, Portugal… Because Data visualization is on FIRE there! Dave got to attend one of their meetups when he was visiting this spring, and two of their leaders really impressed. Sara Mesquita and Caroline Doye lead this awesome group of information designers, meeting almost weekly on different topics from D3.js to Tableau.

Caroline Doye

For Caroline and Sara, they believe that knowing how to extract the information that is "hidden" in the data is what makes all the difference, regardless of your occupation.

The keys to their success are numerous, but something they really focus on is keeping the topics strongly interdisciplinary. They also make sure you walk away with tangible ideas to level up your skills with any tool, from Excel to Tableau or Power BI. Or if you wish to get to the next level using R, Python or even D3.js, this will be at the meetup for you as well.

Community is the key, whether it’s in the United States or in Portugal. It’s useful no matter what level you are. For the dataviz experts, giving back to the community is a way to test and hone their skills. For beginners, it’s a safe environment for people to learn, grow and try new techniques before taking back to a professional setting.

Sara Mesquita

One of the activities that they do that I found fascinating was “data sketching”… putting ideas on paper and trying lots of iterations. Caroline feels that data sketching helps get you beyond just the coding and numbers, but really think about the ideas and try new things.

I think Dave and Matt will be doing some data sketching of our own in the near future!

More about Sara and Caroline

Caroline on LinkedIn - in/caroline-coro
Chantilly on LinkedIn - in/saramesquita1
Data Viz Meetup Lisbon - meetup/Data-Visualization-Lisboa

Permalink

Data Literacy is HOT at Minneanalytics Data Tech 2019

Beyond the Data — Mon, 03 Jun 2019 21:14:14 +0000

For people on the technology side of analytics and data science, there can often be a sense of frustration that the business teams don’t fully understand how to use the data, models, insights, and reports that we create for them. Business teams need to DO SOMETHING with the data, or you won’t see the ROI on your analytics projects.

Matt and Dave checking out the vendor hall at Minneanalytics Data Tech 2019

Matt and I had the opportunity to speak about exactly this topic at the 5th Annual MinneAnalytics Data Tech 2019 conference in the Twin Cities last week. The conference was buzzing with energy this year, with over 1,300 people registered!

We were interested to see how a Data Literacy topic would be received at a conference designed for Data Science and Information Tech people. It turns out that both business and technical people are interested in improving the fundamental data capabilities of their organizations, as 200 people signed up for the event!

Our key message was simple: If you’re an analyst or data scientist, and feel like the business team doesn’t quite “get” what you’re doing… start building relationships with them! Data Literacy for your business teams start with you. They know that data can help them, but you need to bring them along the journey.

To that end, here are our three ideas for Data Science, Analytics and BI teams to start dipping their toes in the water of data literacy for their organization:

Find a business buddy. Get at least one and more preferably business buddies where you can share your knowledge around the power of data literacy. In turn you can get more knowledge around the business domain. Seek out people that are in a similar stage in their career but on business side. important: This is different than and not a mentor relationship.
Do a data viz challenge or hackathon together. Working on the same team in a close time-boxed competitive environment with those that are business-side people will help you empathize and respect them more with them and vice versa. Remember we talk about diversity and it's strength and an element of diverse teams is different organizational backgrounds.
Judge a student data challenge together. There are a lot of student analytics challenges nowadays and they are always looking for judges. Find one and get your business buddy or another business-side person to participate in this challenge as a judge. You will both be better able to understand more of your strengths from the questions you each ask and engagement with student teams. Plus you will be doing a social good by doing this and better helping these students understand different perspectives from an organization.

Thank you MinneAnalytics, their sponsors, and everyone that attended Data Tech 2019 and especially those attending our session. We love coming to these events to see just how powerful a community’s passion around data can be.

Download the presentation

Download

Ep 25 - Brian O'Neill - Designing Better Experiences in Analytics

Beyond the Data — Tue, 14 May 2019 11:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“Good analytics really starts with empathy. It’s truly the heart of good design. It’s caring and having the right conversations with the right people.”

— Brian O'Neill

Design. User Experience. Knowing your audience. Empathizing with your end user.

These are such critical facets of getting analytics right in your organization. If you don’t pay attention to what your constituents want and need, you’ll build something amazing, but it wont’ get the adoption you should. Low adoption means people aren’t using the data you worked so hard to produce.

Data visualization and data storytelling are certainly part of the answer. But it’s only part. A well crafted data viz tells a story, but if nobody wants to read it, then was it really a good viz in the first place?

Brian O'Neill knows a little bit about this problem. He is a product designer and founder of the consultancy, Designing for Analytics, which provides design and UX consulting for custom enterprise data products and apps. For over 20 years, he has worked with companies including DELL/EMC, Tripadvisor, Fidelity, NetApp, MITRE, JP Morgan Chase, ETrade and numerous SAAS startups. Today Brian focuses on helping clients create more useful, usable, profitable, and engaging decision support software and information products. In addition to consulting, Brian is also an international speaker and podcast guest, having appeared at multiple O'Reilly Strata conferences, Predictive Analytics World in Berlin, and on the IBM Analytics podcast, Making Data Simple. He also authored the Designing for Analytics Self-Assessment Guide for Non-Designers, maintains an active mailing list, and hosts the podcast, Experiencing Data. Earlier in 2018, Brian joined the International Institute for Analytics' Expert Network as an advisor on design and UX.

A fun fact about Brian? He’s a musician by training! He is a professional percussionist in Boston. He tours internationally and has performed at Carnegie Hall and The Kennedy Center

More about Brian O’Neill

Connect on LinkedIn - in/brian-oneill-product-designer

Brian’s company - Designing for Analytics

Brian’s Podcast - Experiencing Data

Brian’s Twitter - @rhythmspice

Links and References

Conference - O’Reilly Strata

Conference - Enterprise Data World

Conference - IIA Analytics Symposium

Learn more

Permalink

Ep 24 - Mitchell Grewer - How Cargill Unlocks Data for Everyone

Beyond the Data — Wed, 01 May 2019 12:57:47 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“You have to lower the cost the curiosity. We’ve found more success getting people excited to change, rather than mandating from the top.”

— Mitchell Grewer

Self Service Analytics. It’s been a popular topic of discussion in BI and Analytics circles for quite a few years now. The premise is great. Get more data in the hands of more users. Reporting and Analytics tools like Power BI, Tableau, Domo, Alteryx and Qlik have quickly accelerated the trend and made “non-analyst” data usage more possible than ever.

But is it as easy as buying a tool, giving it to your teams and waiting for the magic to happen?

Of course not. Great Self-Service analytics is hard work. Today we’re talking with one of the leaders in developing this approach, Mitchell Grewer.

Mitchell is responsible for self-service analytics at Cargill in Minneapolis, MN. He has built an amazing energy and enthusiasm for using data to drive their business. But his focus isn’t on hiring more data scientists, or implementing gigantic hadoop clusters. His job is to engage with the business, turning all 80,000 employees into data-wielding mini-analysts.

“I want everyone at Cargill to See and Understand Data. Our goal is to empower all employees to leverage the massive amount of data we have to unlock insights and make better decisions.”

It’s been a long, winding journey for Cargill to get this point, and it certainly didn’t happen over night.

“You have to include both top-down approaches for buy-in as well as bottoms-up. They’re both critical. For us, the bottoms-up was what really helped us take off”.

Mitchell started small, built a community, started training people who were interested, and then launched an enterprise “Data Visualization Challenge”. That challenge is what really got things started. Senior leaders saw what the power of data could really do, and suddenly both analysts and executives were on-board.

Check out the podcast for many other great stories of success, failure, and continued evolution of Mitchell’s self service analytics transformation!

More about Mitchell Grewer

Connect on LinkedIn: /in/mitchellgrewer

Mitchell’s company - Cargill

Twin Cities Tableau User Group - Monthly Meetup

Links and References

Favorite Storyteller or Author - Daniel Arenson

Favorite Blog - Dear Data

Learn more

Permalink

Ep 23 - Renee McGregor - Using Data Literacy to Drive Analytics Adoption

Beyond the Data — Tue, 09 Apr 2019 11:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“Data doubters are so important to our Data Literacy
program... They question everything and they add VALUE to the process”

— Renee McGregor

We're continuing our "Analytics on the Road" series! This week Dave sits down with Renee McGregor from South Africa Qlik, a partner reseller. Over the past couple years, they've really focused on Data Literacy in Cape Town and it shows! If you look up Google Trends results for "Data Literacy", you'll find that South Africa is one of the top countries. Renee talks about what that looks like and the work they do to improve Data Literacy for their organizations.

Renee is a native of Cape Town so she’s seen a huge shift in the culture and the opportunities over the years. She believes that Data Literacy can be that “next wave” of opportunity for organizations and people to change the way they live and work.

So what is Data Literacy mean for Renee? “To read, work with, analyze and argue with data” says Renee. It’s important that people really internalize how data affects them and how they can use data to improve their own lives.

Renee has been solving this for organizations all over South Africa. So what’s the secret? “Adoption at the C-level is absolutely necessary to drive a good data culture”. She’s seen successful implementations and failures and it always comes back to leadership.

The other piece that impacts a Data Literacy initiative in an organization? “You need people who have a passion for data and are ready to share it.” Once you have the C-suite aligned, you’ll need key data champions out there on the front-lines sharing the stories, benefits and skills with everyone.

Thanks for sharing your experiences, Renee!

More about Renee McGregor

Connect with Renee on LinkedIn: /in/rene-mcgregor-a76a0930
Renee’s company - South Africa Qlik

Links and References

Data Hero - Jordan Morrow (Head of Data Literacy @ Qlik)
Favorite Storyteller or Author - Joyce Meyer
Organization - Data Literacy Project

Learn more

Permalink

Ep 22 - Lailah & Julia - Empowering South Africans through Open Data

Beyond the Data — Tue, 02 Apr 2019 17:32:15 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“The government is starting to open some data in South Africa. We need to skill up our citizens on how to access it, how to consume it, and how to use it.”

— Julia Renouprez

First, a quick announcement… We’re launching a new series called “Analytics on the Road”!

For those that don’t know, Dave Mathias has been traveling through Africa and Europe the last 3 months. Along the way, he had the chance to meet up with data people from all walks of life. We wanted to share their voices and find commonality and community, learn from them

He started in Cape Town South Africa where he had the opportunity to speak with Lailah Ryklief and Julia Renouprez from the non-profit organization, Open Up.

At Open Up, they believe that an equal society starts with equal access to information, and that access to relevant information creates an active citizenry. You cannot change something if you don't know what it is, how it works, or that it even exists. Open Up is helping the citizens of South Africa by working with the government to open source their data, and then building tools and skills to make that data available and understandable to the broader population!

One of the core ways they help citizens in South Africa is through Data Literacy training, helping people understand that the data exists, what kinds of questions they can answer with it, and how to use it through some simple tools.

South Africa has a lot of opportunities for growth. One of the biggest problems they face right now is poverty. Helping people understand where poverty comes from and how it is perpetuated is the first step in reducing the poverty gap.

Teaching people about data also gives them critical skills that they can use to find better jobs, increase their pay, and improve their and their families lives!

More about Lailah & Julia

Connect with Lailah on LinkedIn: in/lailah-ryklief-b6489054
Read Lailah’s Bio: Lailah Ryklief Bio
Connect with Julia on LInkedIn: in/julia-renouprez-9ba25619
Read Julia’s Bio: Julia Renouprez Bio

Links and References

Lailah & Julia’s Non-Profit - Open Up
Julia’s Data Hero - Adi Eyal - Open Up Director
Lailah’s Favorite Book - Forensic Architecture
Julia’s Favorite Book - The Witcher books
Cape Town Communities - Codebridge Meetup
Cape Town Communities - Data Viz Meetup

Learn more

Permalink

Ep 21 - Dennis Still - Analytics for Startups and Small Business

Beyond the Data — Tue, 26 Mar 2019 11:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“Big data, small data, medium size data. The real question is what you’re going to do with it once you have it.”

— Dennis Still

Dennis Still is a startup analytics expert. Where many analytics projects come from large $1B+ organizations, there are many startups and small businesses that need similar capabilities. How do small business compete and keep up? What kinds of challenges do these companies face, and how do they view their own relationship with data?

Dennis has had a wide and varied career that didn’t necessarily start in data, but has taken him on a journey through various startups like When I Work, Gov Delivery, and his very own startup, Bigfoot Analytics.

Dennis has spent years working directly with entrepreneurs, figuring out what they need from their data, and delivering insights that drive exponential growth as they disrupt their various industries.

One of the biggest things he found through his time leading data and analytics, was that the questions that small companies are asking aren’t that different from what big companies are asking, just on a different scale. Revenue, costs, customer satisfaction. The key to his success was in being able to help his C-suite leaders identify their KPIs that would fuel rapid growth.

“Our CEO would throw out a number and 85% of the people in the room knew what that meant. It was the number that we used to drive the business forward”.

Despite being more nimble, Dennis feels like there’s a lot of opportunity still. "Who owns the data? How does it work? Who is going to look at it and do something with it?” A data-informed culture doesn’t just happen. It takes work from the analysts consistently pushing the metrics and getting the leaders to embed it into their communications.

More about Dennis Still

Connect on LinkedIn: in/dennis-still

Links and References

Dennis’ Company - Bigfoot Analytics
Data Hero - Kirk Borne (Twitter @KirkDBorne)
Data Hero - Carla Gentry (Twitter @Data_Nerd)
Favorite Blog - Jeffalytics (Google Analytics Consulting)
Favorite Podcast - Digital Measure Show w/ Brian Poe

Learn more

Permalink

Ep 20 - Serena Roberts - Authentic Relationships That Build Analytics Success

Beyond the Data — Tue, 19 Mar 2019 11:30:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“Your job does not end when the analytics development is done. That’s when it starts.”

— Serena Roberts

Serena Roberts is a force to be reckoned with. She’s a mom, an analytics leader, a Tableau ambassador, and the driving force for two great local communities that she runs. That’s why we were so thrilled to get a few minutes of her time to talk about her approach to data, and how she’s become such a successful data leader.

Like many of us, Serena didn’t start out on a path to be a “data person”. She kind of fell into it by accident… A happy accident that’s taken her on an amazing journey from “business professional” to data guru, to fearless leader of the Twin Cities Tableau User Group, and the Minnesota chapter of the organization, She Talks Data.

The first thing that you’ll notice about Serena when you talk to her is her passion for understanding other’s needs. She deeply cares about the people in her life, both professionally and personally. She’s also a data visionary. She sees where her organization needs to go and isn’t afraid to communicate that vision to others, even when met with resistance.

Serena talked to us about her time at Capella, and reflected on some of the lessons learned while working for a well-established and relatively “analytics-averse” organization. “I was the unpopular person pushing new ideas”. “We were really trying to change human behavior, one sales rep at a time”.

“I saw such a need for analytics and data that no one was looking at. I was the unpopular person pushing new ideas and challenging the status quo””

The more you talk to Serena, the more you realize that it’s HUMANS that drive her, not the DATA. For her, it’s all about helping individuals use data in their day-to-day jobs. Enabling those fundamental data literacy skills that empower a business person to make a better decision, communicate more effectively, or answer a question faster.

“People focus on technology and process… and they heavily underestimate the change management aspect of the work we do as analysts.”

Change management is about changing human behavior. That’s the hard part. But the data community is lucky to have people like Serena driving their organizations to a more data-informed culture.

More about Serena Roberts

Connect on LinkedIn: in/serenaroberts

Links and References

Meetup - She Talks Data
Meetup - Twin Cities Tableau User Group
Book - Rebels at Work - Leading Change from Within
Book - Peanut Butter and Cupcake
Serena’s Favorite People - Nick Peterson, Karla Hillier, Sherri Benzelock, Laura Madsen
Award Winning Tableau Team - Jeff Plattner, Tricia Duncan

Learn more

Permalink

Jeff Sloan - Better Analytics through Product Management Principles

Beyond the Data — Tue, 12 Mar 2019 11:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“The best analysis for YOU might not be the best analysis for your USER. Make sure you think about the outcomes that you’re trying to drive.”

— Jeff Sloan

Today, on the podcast we’re trying something a little different. For those who may not know, Dave has been travelling through Africa and Europe the past 2.5 months as a part of a program called Remote Year.

During his time in Cape Town South Africa, Dave was introduced to our next guest, Jeff Sloan who is also a part of Remote Year. Jeff is a self-prescribed Data Product Manager and they thought it would be fun to try recording some of their conversations.

They headed over to a local coffee shop and set up the mic and started chatting about data, business intelligence, product management and how these disciplines are starting to intersect.

If you hear banging dishes or cars driving by, feel free to imagine sitting outside sipping latte’s on a lovely warm day in an open-air coffee shop in downtown Cape Town, South Africa.

So, what is a “Data Product Manager”, exactly?

According to Jeff, this role is responsible for thinking about the data infrastructure of the entire organization, mapping out the flows, sources, and storage platforms for both internal and external data. It’s the first step in empowering things like Machine Learning, AI, and A/B testing across the organization.

Jeff loves data, but he feels like many of the traditional ways that organizations use data and analytics today could be even better. That’s why he’s so interested in bringing product management concepts to data to deliver more value and better, more integrated insights. Product management tools like SCRUM, backlog grooming, and user experience can all play a role in driving more value.

“It’s important to understand where your [internal or external] customers are coming from. The more we can understand what they need and how they need it, the better we will be as data people… We need to take them along on this journey to using data.” How do we do this? Jeff recommends starting with the business question and then working backwards from there, in an iterative and agile way. This will ensure that the insight/analytics produced meets the needs.

Thanks for sitting down for coffee, Jeff. And good luck on the rest of your Remote Year travels!

More about Jeff Sloan

Connect on LinkedIn: in/jeffreymsloan

Links and References

Slack - DBT Slack Community (DBT = Data Built Tools)
Meetup - London DBT Meetup Group
Consulting Group - Fishtown Analytics
Book - The Secret History by Donna Tartt
Book - The Coaching Habits by Michael Bungay Stanier
Blog - Locally Optimistic
Jeff’s Company - eMoove UK Online Real Estate Agent

Learn more

Permalink

5 Things Executives Must do to Support a Data Culture

Beyond the Data — Wed, 06 Mar 2019 12:00:00 +0000

Executives today are faced with many pressing needs. Customer success, your P&L, internal politics, shareholders and investors, managing your teams, strategy and goals, keeping your key projects moving forward. But there’s a new growing factor to add to this never-ending list: Data.

You hear about it from Harvard Business Review. You hear about it from Forbes. You hear about it from McKinsey, and the Wall Street Journal, and CIO. You hear about your competitors doing things with data to get an edge.

In a previous article, we talked about the critical components of making data work in your organization. Hint: It isn’t just investing in a data science team and then waiting for profits to roll in.

Culture change must be part of the equation. Probably not what you wanted to hear, but that’s what it’s going to take. Changing your team’s culture takes several things, but the critical part we’re discussing today is the top-down approach.

Here are the five starting items that executives need to consider when implementing your data culture change.

Support a data-informed decisioning culture

This is first because without this then there simply isn’t anything else. Everyone in your organization must be on board to seek out data, learn from data, and make decisions based off analysis. A core tenant of hiring, promoting, and rewarding people needs to be off of strong data-informed decisioning. This applies just as much to executives themselves as their staff. Too often data is produced to back a gut-based decision and proper analysis and experimentation not performed. Then, when something does not workout then people raise their hands saying the data told them to do so but instead data just supported the desired outcome. There is a component to this item which requires that data and ability to access and analyze it must be put in place and maintained through proper data governance and self-service business intelligence platforms.

Support your Key data champions

Every organization needs data champions to keep your momentum going. You should have many of these data champions, embedded in the business lines, singing the praises of analytics and what data can do for them. Who are your data champions? Are they being recognized, rewarded, and empowered in their efforts? It is vital that executives understand that data champions are needed to drive data culture bottom-up.

Support data-informed decisioning technologies

It is no surprise that having appropriate data and analytics technologies available for not just the analytics teams but also the business teams is a must. Having the proper tools to do the job whether it is Tableau, Qlik, Power BI, Domo and others. That data and ability to access and analyze it must be put in place and maintained through proper data governance and self-service business intelligence platforms.

Support an information-sharing culture

It is not alright for departments to silo off data so they can benefit from it and other departments can’t. Yes, there are instances that data cannot be shared for various data privacy reasons. But, when data is shareable within an organization, the default should be to do it. It is not alright for departments and people to indiscriminately put up data silos against other areas of the company.

Support organization-wide data literacy

Data literacy is essential for all of your employees. This doesn’t mean that everyone needs to be a data scientist. In fact, there are different levels of data literacy needs depending on organizational roles. First, understanding your employees’ data literacy is essential. Then, helping those employees close data literacy gaps with training that is done in an engaging and practical way.

All these items are essential for executives to drive a data culture. However, it is really important to point out that executives must eat their own dog food. No longer is it alright for you to tell others to do what you say, now what you do. Demanding data in your own decisions and even getting hands-on with an executive level dashboard should be expected.

Taking these items and putting into practice will help create a data culture at your organization. Then, everyone will not only be speaking the language of data together and making decisions on analysis in a sustaining data culture.

This article appears in a series of blog posts about Data Culture, Data Literacy, and why it matters for organizations to think beyond Data Science. If you liked this article, make sure to read the rest of the series:

Five Reasons Why Data Culture is Just as Important as Data Science

The Key Roles of a Data-Informed Organization

Who is Driving your Data Culture? The Role of the Data Champion

Jordan Morrow - What Data Literacy can do for you

Beyond the Data — Tue, 05 Mar 2019 12:00:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“You can’t just throw data like spaghetti against the wall and see what sticks. Think about outcomes you want to acheive”

— Jordan Morrow

Jordan Morrow is on a mission to help organizations and individuals become data literate. He believes that the ability to speak, read and write data will be the next big differentiator in the next few years. He travels around the world speaking with people about how to improve their own data literacy.

Jordan is also just really fun to talk with. We ended up chatting for a long time (we edited it down a bit for your sake) about Data Literacy and how it fits into Data strategy, Data culture, Data science, and how it can drive real, tangible outcomes for organizations.

We also talked about a real-life example of what Data literacy and good Data culture looks like. The Avon Somerset Police Force is enabling their staff and officers with data, helping them understand how to interpret the results and what to do with it, and it’s having a real, positive impact on how they do their jobs!

We asked Jordan how an organization like that could get to a point where they were changing culture. It boils down to three things:

1) It starts with a leader who understood the value of data

2) It requires training. Not just for analysts sitting back at the home office, but for the officers out patrolling the streets each day

3) It requires communication, roll-out and adoption plans to ensure the culture change “sticks”

We talked a lot about an outcome-based model to make data truly powerful. Let’s start with what we want to achieve… “We want more sales”, “We want more return on equity”, “We want higher employee engagement”. Let’s start there and then bring data to the table to help solve that. The worst thing we can do is use data to confirm our own biases.

One of the cool projects that Jordan works on is the Data Literacy Project. As the board chair, he started this project as another way to help people and organizations become more capable with data (Dare we say… Data Able?). It’s a fantastic source of stories and tools to help YOUR organization get started.

We had so much fun talking with Jordan and can’t wait to have more chats in the future!

Thanks so much for coming on the show!

More about Jordan Morrow

Connect with Jordan on LinkedIn: in/JordanMorrow

Check out the Data Literacy Project: TheDataLiteracyProject.org

Follow Jordan on Twitter: @Analytics_Time

Links and References

Book - Can’t Hurt Me: Master Your Mind and Defy the Odds by David Goggins [Explicit]

Book - Freakonomics by Stephen Dubner and Steven Levitt

Podcast - Data Skeptic by Kyle Polich

Podcast - Joe Rogan Podcast [Explicit]

Podcast - Revisionist History by Malcolm Gladwell

Blog - Qlik’s Data Literacy Blog

Learn more

Permalink

Building a Data-cated Community with Kate Strachnyi

Beyond the Data — Tue, 26 Feb 2019 12:30:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

Kate Strachnyi is a superhero.

Seriously.

She writes books, she works a full time job, she hosts the Humans of Data Science video channel, she’s a Udemy instructor for Tableau, she started the Datacated Weekly project, and she still has time leftover to be a mom and to run crazy-long marathons!

Dave and I were thrilled to sit down with Kate for a few minutes and learn about how she got into analytics and data science. Hint, it wasn’t her original career path!

We also learned a little about her role at her Big 4 consulting firm, where she works on executive reporting and analytics for the C-suite. She also helps drive Tableau and Power BI Self-service adoption across the different business teams. Her keys to getting people on board? Start with leadership buy-in and then simply show people the power of the tools. Software like Tableau and Power BI make it easy for non-technical users to jump in and start using their data.

“Show people that dashboards are not that hard to build. Once they see that there’s not much friction to get started, they’ll start using it.”

— Kate Strachnyi

Kate has a huge following on LinkedIn and Twitter (for good reason). We asked her about the community she’s built around data. She says that the people side of data is the most interesting part of being in our industry. Learning and growing together is far more interesting than trying to do it alone.

And boy has she done just that. And she encourages you to create your OWN community! As we wrapped up our time together, she shared her step-by-step process for creating your own analytics or data science public project to start sharing insights and learning from others. You’ll be amazed at what happens when you do!

Thanks so much for coming on the show, Kate!

More about Kate

Connect with Kate on LinkedIn: in/kate-strachnyi-data

Check out Kate’s Website: storybydata.com

Follow Kate on Twitter: @StoryByData

Links and References

Udemy - Tableau Visual Best Practices: Go from Good to GREAT!

Book - Journey to Data Science

Book - The Disruptors: Data Science Leaders

Cathy O’Neil - @mathbabedotorg

Book - Weapons of Math Destruction by Cathy O’Neil

Book - Lean In by Sheryl Sandberg

Book - Extreme Ownership by Jocko Willink & Leif Babin

Learn more

When Is it Okay to Ignore the Data?

Beyond the Data — Tue, 19 Feb 2019 12:30:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

“Data is there to answer a question. But human intuition needs to play a role as well. Combining the two is the goal.”

There is no easy button for determining how to use your data. We like to think that data is this perfect, impartial mediator for human emotions and bad decisions. But really, the data is just a full of biases as your intuition and “gut’ is.

An analysts job is to understand what types of biases could potentially impact the output of your analysis, dashboard or model, and then ensure that the data users know the pros and cons of your dataset.

In this episode of Data Able, we’ll talk about looking at your datasets, making judgement calls about that data, and questions to ask of your data’s origin/source. We’ll also cover some real-life examples from our own pasts where data we used were suspect and how we handled them.

Most importantly, we’ll talk about some strategies for managing through the inherent bias in your organization’s data, and how the “manage” your end-users through that process. Getting executive buy-in and ensuring everyone is comfortable with the pros/cons of the dataset BEFORE you deliver the analysis is the key!

Links and References

Understanding Data Governance - CIO.com

Understanding the Types of Data and How They’re Captured - HBR.org

Learn more

Permalink

Improving Higher Education Through Data with David Niemi

Beyond the Data — Tue, 12 Feb 2019 12:30:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

David Niemi loves higher education. So much so that he’s spent his entire career involved in it. From an early age, David recognized that there were better ways to help students and learners achieve their goals, and he’s been on a mission to make that experience better ever since. Throughout his career, he’s been a teacher, student, EdTech leader, professor, and analyst. David perfectly straddles the line between technology, data science and education, which makes him well suited for leading Kaplan’s Learning Analytics division, as the VP of Measurement and Evaluation.

EdTech has come a long way in the last 20 years. But even today, David believes there’s lots of opportunity to do it better. He starts with a basic question: “If we actually built ed-tech that taught people something, how would we know if they’re actually learning anything?”

This is the foundation for David’s role at Kaplan. He’s looking past “completion rates” and “GPA” and looking at measuring the real skills that are transferred to the students. He’s focused on the learner outcomes, like getting a job, increasing their salary, and improving their lives and communities.

So what are the key metrics or questions should Higher Ed be focusing on? David boils it down to three easy points: Are the students learning something? What is the level of student engagement during the course? What are the measures of student motivation throughout the course? These are different than the typical metrics because they are collected in near-real-time and provide teachers with tailored feedback on each student that ensures they’re getting the right level of instruction at the right time.

“A measure of learning should tell you what new skills, knowledge, ideas and concepts have you developed. Not how many courses you completed.”

— David Niemi

David also shared some interesting correlations between how to successfully educate learners and how to run successful analytics projects. In both cases, you need start with the end in mind… For education, it’s

1) what do you want to do in your career?

2) What skills do you need to get there?

3) Which classes or programs will provide those skills?

This is exactly how analytics projects should work!

1) What does the business need to solve?

2) What data do we need to inform those decisions?

3) What techniques do we use to tease the answers out of the data?

We also talked a bit about David’s new book, Learning Analytics in Education which is a set of research studies focused on pairing education data with data science techniques to drive better engagement for students, whether in online classes or in-person.

The book is one of the first to look at these new EdTech platforms that allow for ongoing measurements of student progress. They investigate how they can use these new data points to help educators increase their students’ success. These educators can now harness data to personalize the experiences for learners, while improving overall outcomes at scale.

If you’re at all interested in this brand new space, we strongly encourage you to pick up a copy!

And thanks to David for coming on the show!

More about David

Check out David’s book on Amazon: Learning Analytics in Education

Connect with David on LinkedIn: in/david-niemi-2630757

Follow Kaplan on Twitter: @KaplanNews

Learn more

Permalink

Who is Driving Your Data Culture Transformation

Beyond the Data — Wed, 06 Feb 2019 19:34:35 +0000

There are several critical roles that are critical to increasing the maturity of your analytics. But the glue that holds it all together is the person we refer to as the Data Champion. You won’t see a job description for a “Data Champion”, but all organizations that have a strong data culture will have at least one, and likely more.

Data Champions are people who spearhead data culture within an organization. Sometimes, they are an executive or senior leader, but oftentimes, they’re the boots-on-the-ground people who are simply passionate about promoting and improving data-informed decisions for their organization. They may be part of a business team, data team, or technology team. They may be extroverts or introverts.

What does a Data Champion look like?

Data Champions are natural disruptors, communicators, and networkers who can establish, drive, and support a clear, data-informed vision. Data Champions are made, not born. You will often spot them because they will be seeking to start an internal meetup around a data-related topic or starting a data visualization competition, or maybe they will be the person who is at another’s desk showing them how to approach a data problem. They aren’t necessarily the most technical person in the room. But they are most certainly the ones who are building communities, telling stories about the possibilities, and focused on embedding analytics into every corner of the organization.

These Data Champions will be present in an organization whether they have been sought out or not. Organizations with a strong Data Culture, though, will have more of these Data Champions, and their level of empowerment and satisfaction will be higher.

What does a Data Champion do?

Data Champions play a key role in helping translate between the business and their area of the organization to help drive data usage when making decisions. They engage with business and technology partners to ensure they are smoothly working together. Further, Data Champions will have relationships with other current and future Data Champions, including those not within the Chief Data Officer’s direct area.

Data Champions are more than just translators though. They create vision, alignment, and empowerment for the teams they support. They build energy and excitement for a data-informed approach. They are skilled at working with business leaders to build trust in the analytics solutions being built. They constantly communicate the benefits that data can provide and the results that the organization has gotten from analytics investments, and they communicate the vision for the future.

Champions are Critical but not sufficient

Getting the organization moving in the right direction is obviously important. However, doing so without executive buy-in will result in frustration, limited results, and a lack of funding. Executives have to be part of the equation.

Similarly, moving forward without a technology foundation (quality data, storage platforms, reporting tools), and skilled analysts to dive into that data, will also result in limited results and frustration. The data team and technology must be a critical part of the equation.

Finally, it’s important to note that the best champions are the ones who work themselves out of jobs. “Translating” between the business and analyst teams is critical in the early going. But think of the benefits of translating; it didn’t need to happen, and both teams simply spoke the same language. Reduced friction, reduced effort, and faster/clearer communication would result. The data champion only translates until they can get the teams talking in the same “language”.

Here’s a great video about how Data Translators are critical pieces but are a stepping stone to the whole organization being data literate.

Credit: McKinsey & Company

The Key Roles of a Data-Informed Organization

Beyond the Data — Mon, 04 Feb 2019 19:08:00 +0000

Most people would all agree that data is a key go-forward strategy for their organizations.

As we discussed in a recent article however, there are some significant challenges that come with executing on that strategy. How do we overcome those? You need to embed analytics and data science directly into your organization’s culture.

There are three interlocking roles, each with some level of responsibility for making analytics work. The fourth and most critical role, the “Champion” sits at the center of these roles, driving alignment between everyone and driving successful change managment.

Over the next couple weeks, we’ll break down these roles in much more detail, but here’s a high-level overview:

The Executive Team

The CEO, CFO, CMO, CHRO, an the rest of the C-suite. When push comes to shove they need to support data initiatives, support the financial investment in the, weave data into the strategies of the organization, and ultimately hold the organization accountable to data-informed decisions and actions.

The Business Team

The many core functional areas of your organization. From human resources to sales to product to finance, the business team is critical to driving successful analytics. They must be on board and empowered to use data. Without this team informed, engaged and comfortable with data, then your amazing analytics outputs will fall on deaf ears, and the potential business value will be lost.

The Data Team

The extremely adept technical team who will be moving, storing, touching, analyzing, manipulating, and communicating your data. There are many roles within this broad category, but could include people like BI Developers, ETL Developers, Business Analysts, Data Scientists, and Report Creators. The key to their success is to turn them into key business partners, rather than basic order takers.

The Data Champion

The highly driven person or team at the center of it all. They are the evangelists that shout from the rooftops the importance of data for your organization. They “translate” how data can help the business, communicate it to leadership, and ensure the data team executes on the efforts. Data Champions are natural disruptors, communicators and networkers who can establish a clear data-informed vision. They create excitement and energy around data, and know how to influence the other three groups on how to execute.

These stakeholders together provide the pillars of support for an organization’s data culture. If one or more pillar is out of alignment, then the whole data culture is weakened. One pillar is not more or less important than any other. They each play a role in driving the data maturity of the organization and in-turn, the value that can be captured by analytics.

So what about your organization? Can you identify the people who fall into each of these groups? Are each of them in alignment with each other? What is the missing link that is holding your organization back from leveraging data effectively?

Need help?

Contact me?

Stories from the Data Trenches with Liz Weber and Tessa Enns

Beyond the Data — Tue, 29 Jan 2019 12:30:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

Back in December, Dave and I had the privilege of doing a speaking event at the MinneFRAMA (Finance, Retail, Marketing Analytics) event. Naturally, we decided to try something different and tape a live episode with two of my favorite analytics professionals in the Twin Cities, Tessa Enns and Liz Weber.

It was truly a once-in-a-lifetime experience: Sipping coffee, and talking analytics with these two amazing women. The venue didn’t hurt either! We were in a huge room at the Minnesota Science Museum, with our backs against a wall of windows overlooking the Mississippi river. We learned a lot about how to make sure your analytics projects are truly successful.

Tessa Enns

“I’m not working to fuel my technical skills, my technical skills are working to fuel the business challenges”

— Tessa Enns

Tessa talked to us about “accidentally” coming into an analyst role at Cargill, being given a huge transportation dataset and being tasked with finding something in it. Tessa is the kind of amazing person who looked at this as an opportunity, and went right to work, learning the data, learning the business, and learning technical skills along the way.

What is so amazing about her journey is that she was able to build a strong relationship with the business, who now trust the data, find opportunities to improve, and know how to turn the numbers into action that drives real monetary value.

Tessa preaches an approach where analysts need to “lead with the needs, not with the data”. She says this helps the analyst understand the real problem and help solve it. She also recommends putting every insight into dollar terms that your business will understand. “I put the cost savings or cost impacts right at the top of every dashboard”.

Liz Weber

“Take some time to understand the business processes that create your data. You will be able to understand your business teams and tell better stories that drive change”

— Liz Weber

Liz talked to us about a highly complex pricing challenge that her VP faced a number of years ago. The team was going through a major transition, and had invested a huge amount of money in their business. The VP wanted a dashboard to start monitoring whether this transformation was going successfully.

Liz started with the end in mind. Learn about what the leader wants, where they’re trying to go, and what the key measures of success really look like. She then sat down with the IT team, and the business teams underneath this VP. Making sure that everyone had a voice in the project was mission-critical to make sure that she 1) had the right resources, 2) had everyone moving in the same direction, 3) made the solution better than just her own ideas. It also helped with adoption, and making sure that everyone actually USED the end product.

What she learned from this project was that your leader/sponsor matters. If you don’t have the right sponsorship, it doesn’t matter how smart you are, or what kinds of technical knowledge you possess. You need to make sure you’re aligned to leaders who are committed to doing something with the outputs you produce.

Resources & Links

Learn more

Permalink

Five reasons why Data Culture is just as important as Data Science

Beyond the Data — Wed, 23 Jan 2019 12:30:00 +0000

Data Science is an amazing tool in the organization’s toolbox. It can provide immeasurable value when done well. Those in data circles have spent the last 10 years hearing about those success stories, reading example after example of the great things it can create. To be sure, there is still untapped value in that “oil well”.

And yet, while executives have been steadily investing more resources into tapping this seemingly endless well of value, we’re starting to see the cracks. Gartner estimates that by 2022, 20 percent of analytic insights will deliver business outcomes. Wait, what? If that’s the future, what does the current state look like?

Anecdotally, you see it too. The business is frustrated that they’re not seeing the returns. The CFO is starting to scrutinize those budget lines closer. The CEO is getting more impatient for results.

Is this a failure of the data scientists? Maybe we couldn’t afford the “good” ones. Perhaps they didn’t do enough data visualization or data storytelling?

Perhaps it’s a failure of the executives? They didn’t invest as much as they needed to get the return. Or, perhaps they weren’t fully bought into this new approach.

Perhaps it’s a failure of the business teams? They didn’t value what the data scientist could bring to the table. Or, perhaps they didn’t listen to the recommendations and insights being generated.

Or worse yet, data science just doesn’t provide the value we thought it did. Yes, people are actually asking this question.

Data is not a siloed activity

Separating the data teams from the business teams is a surefire way to never get value from your data

I once worked with a brilliant analyst named Dan. He was a true data scientist before that term was popularized. A truly skilled mathematician, gifted with just about any statistical tool you could give him, he could quickly diagnose and solve the stickiest data challenges we had. I learned a lot from Dan early in my career.

But Dan wanted nothing to do with the business team he supported. Meetings with the business partners was a nightmare for him. He was clearly intellectually superior to those regular old business people. He knew what they needed and he was going to provide them with that. Besides, he was on the analytics team and didn’t report to them, so it didn’t really matter what they wanted.

Perhaps not quite as extreme as Dan, but I often come across this “Analytics vs. Business” mentality in analytics teams that I work with. And I think the root of the problem is that we’ve structured our teams in such a way that we’ve isolated them from each other.

How many of your organizations set out to build more analytic capabilities by hiring a “head of” analytics, then building out a team, then going to work building (or re-building) the data infrastructure? I’ve personally worked with a dozen Fortune 500’s that fit this bill. It’s not a bad place for a startup analytics group. Silo your efforts so you can focus on laying the groundwork. Afterall, you can’t extract value out of your data if you aren’t correctly capturing, moving and storing that data.

Unfortunately, this approach has a sinister downside. It creates an isolated bubble of data-related activities and projects. And the longer you stay in the bubble, the harder it is to push beyond that bubble.

Great data projects don’t happen because isolated data scientists are casually strolling through the data, looking for interesting tidbits. The best data scientists I know today build strong bridges between their team and the business team they support. They understand that the data is there to augment the business, and isn’t there as an end-all-be-all panacea.

Algorithms don’t add value, people do

The most accurate forecast model ever created adds no value if the business doesn’t do something with it

A number of years ago, I was working for a company that was struggling. Sales were down, costs were up, and the industry was in the middle of a seismic shift. The company was caught in a bad situation and was unprepared to deal with it. To make matters worse, we had purchased several other struggling competitors as a way to “bring consolidation” to the market, but all it did was increase the number of low-revenue/high cost problems we needed to solve.

My data science team decided it was time to start leveraging our data to create some actionable tools that could start moving us in a better direction. We set to work building a complex algorithm that would create a series of benchmarks around our customer’s buying history, compare each customer against the benchmark, and expose each customer’s unique “hidden opportunities”. We reasoned that mining over 1 Billion records to find nuanced, customer-specific buying patterns would arm our sales team with critical insights about their clients that would help them target their conversations and drive sales. It was a brilliant plan.

How much increased revenue did it generate? None. We tweaked the algorithm for almost 3 months, adding depth, complexity and accuracy. Getting it “just right” was important. Then we gave it to the sales team who… did nothing. It was too abstract, too complicated. They poked holes in calculations. In the end, it never was rolled out to the sales team, and 3 months of great Data Science efforts were wasted.

6-8 years ago, the industry told everyone that Data Science was the answer. It told employees to rush out and re-brand their skills, or train on those skills. Data Science Bootcamps were created, Universities built new Masters programs. Coursera and Udemy exploded with online certifications. Don’t get me wrong, these are good things! Great things, even.

But what was missing from the R and Python training, the Hive, Spark, Tensorflow, and AI training, the Bayesian probabilities and the clustering techniques… was the reason we create these things in the first place. It’s not about the algorithms. It’s about the human behaviors that our data influences. That’s when data becomes valuable. If we’re not creating things for humans, with the goal to influence those humans in some meaningful way, then our fanciest and most sophisticated algorithms won’t add a drop of value.

Knowing what to focus on

Data can solve problems, but knowing which problems to solve is the real question

Erin is the VP of Sales at your organization. Sales are flat and Erin is growing increasingly anxious about getting back to growth. You decide to take a look at the customer sales data. Is the miss coming from a certain product line? Is it a customer retention problem? Are new business efforts lower than expected? Perhaps deal sizes are shrinking? After some digging, you find that most of your sales are doing fine, but the annual revenue generated by 8 out of your top 10 customers has shrunk over the past 12 months. Erin is now armed with just the information she needs to address the problem and get the company moving in the right direction again.

But what if Sales are booming? Erin is likely just trying to stay above water with all the new deals. She has plenty of problems, but they’re quite different. Now she’s worried about how the Operations team is going to handle the influx of new orders. She needs to make sure that legal has time to review all the contracts. She’s looking into hiring new sales staff to keep up with demand. If you come to Erin with a customer analysis and breakdown of where she’s missing, it probably won’t be received well. She’s got too many problems already and doesn’t have time to fix this problem too. Erin doesn’t need NEW problems, she needs to solve NOW problems.

The point is, you could do the same analysis in both situations, and receive a drastically different response from the leader. That’s because data is only useful when it is being used to solve problems that your stakeholders care about. When there’s misalignment between the business and the data teams, you miss huge opportunities to leverage data.

Business teams don’t understand data

Don’t turn your marketing team into Analysts… arm them with data to become even better at their jobs

Do you have that special person in your life that’s still slightly scared of their computer? Like it’s going to eat them if they turn it on? This is exactly how my grandmother feels. We finally got her to log into her email account once a week, but it took 3 years of convincing.

My grandma is a brilliant lady. And yet she still spends an enormous amount of time with basic tasks like maintaining her calendar, managing daily tasks, basic communication. The computer is there to help her, yet she keeps doing things the way she knows. She’s COMFORTABLE with her old process. She knows it, and it hasn’t failed her so far.

I would argue that most business lines have a similar comfortability with their own industry knowledge. It’s what has gotten them to this point in their careers, and they’ll be damned if a data person comes in here and tells them something different. It’s not that they think data is bad, or doesn’t add value. In fact, if you asked them, I bet 9 out of 10 would say that they need their business line to be more analytical.

But saying you should do something, and actually doing it are very different things. What is stopping them from taking a data-informed approach? It’s fear of what they don’t know. Using data gives up control and safety of their industry expertise for something foreign, confusing, and different.

The lesson here is that getting your organization to use data isn’t about better algorithms, more hadoop clusters, or even more dashboards. It’s about the business becoming comfortable with using the data they have and blending it seamlessly with what they already know.

Executives haven’t truly bought in

executives need to live it the data-driven mindset, building it into their plans, and bringing their teams along

You’d be hard pressed to find an executive that hasn’t added data and data science to their list of yearly goals. They’ve read enough books, seen enough articles, and heard their colleagues and competitors talk about it to know that it can help them. So they dip their toes in the water… perhaps a bullet point at the end of their annual roadmap deck that says “oh yeah, and one more thing… we need to be data-driven”.

Wow a whole bullet point!? What a show of commitment! I’m sure your teams will jump right on it. I mean, they have no training in this, they don’t really know what “data-driven” even means, and they likely have a 10 other areas of focus coming out of that roadmap deck. But yeah, I’m sure it will get done.

Joking aside, most executives have likely allocated a bit of resources, but they’re not fully committed to the idea. To them, it’s still a nice thought experiment. But this is exactly the problem. It signals to their teams that they don’t need to take data seriously. A passing fad.

The unfortunate side-effect of leaders who keep data at an arm’s length, is that when the going gets rough, they are unlikely to rely on it in that critical moment. And they’re even less likely to ensure their teams rely on it. They’ll revert back to their guts. There will be “reasons” why we can’t rely on the numbers, especially when the numbers aren’t stellar. You will hear things like “You didn’t consider that our biggest customer doesn’t go through that [data collection] system”. Or, perhaps “we already tried analyzing that data and it was inconclusive”. Or, the worst “we already knew this information”.

The point is that an organization must have leadership on board with a data-informed culture. They can make or break whether the organization captures the value it’s seeking. Make sure your leaders are actually on the train, not just punching a ticket.

So What is Data Culture?

What is the solution to Data Science’s woes? A more HUMAN approach to data. One where the focus is on alignment between the executive teams, the business teams, and the data teams.

You want to be successful with data?

Do the hard work to ingrain it into the core DNA of your organization.

It needs to permeate how each employee thinks, that they have a voice in their heads asking “what does the data tell us about this?”.

When the business has a fundamental understanding of data, it allows them to speak a common language, often referred to as Data Literacy. This common language builds trust and encourages collaboration between the business team and the data scientists. It opens up the opportunity for them to ask bigger, more impactful questions because they know that they can even attempt to ask them. It ensures that they are more comfortable with a sophisticated mathematical solution to their problem, even if they don’t fully understand it.

But most importantly, it allows the business to bring data to the table, combine it with their deep domain expertise, and make an EVEN BETTER decision than they would have otherwise.

Want to use data more effectively? Align your data science teams with leaders and business teams to make sure they’re all moving in the same direction, and basically aware of each other’s needs and capabilities.

Create a culture of data, help your business team “speak the language of data”, and make sure the data team is tightly aligned with executive & business team objectives. Do this, and you’re all but guaranteed to see data project success rates well above 20%.

The Importance of Data Storytelling Pt 1

Beyond the Data — Tue, 22 Jan 2019 12:30:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Stitcher Pkt Cast

Episode Summary

Storytelling has a long history and is one of our most basic ways to pass along knowledge.

The tradition of storytelling to pass along knowledge and inspire dates back thousands of years. Much of our history storytelling was the main means of passing along knowledge. Even though we have fancy technology though storytelling is just as important today. We are inspired by leaders and orators that can tell engaging stories.

This episode is the first of a two-part series on the importance of storytelling. In this episode we discuss why storytelling is so important. Facts are important, but human emotions are even more important. Simply putting facts on a page won’t necessarily elicit a change in a person. Storytelling helps a person relate to the information you’re trying to communicate.

How about a few tips for people to practice when storytelling? First, know where your story is going, and be able to summarize what the point is. “What is the moral of the story”? Second, re-framing the story into something that people understand. So rather than stating a bunch of generic numbers about how many items move through your supply chain, tell a story about a bag of frozen peas, and how it got from processing facility to your kitchen table.

In Part 2 of the Data Storytelling series, we’ll discuss more tips and tricks on effective data storytelling.

Resources and Links

Some great resources that can help you get started around storytelling include:

Data Viz Made Simple with Kristen Sosulski

Beyond the Data — Wed, 16 Jan 2019 12:30:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Pkt Cast Stitcher

Episode Summary

Meet Kristen Sosulski. She’s an Associate Professor of Information Systems and the Director of Learning Sciences for the W.R. Berkley Innovation Lab at New York University’s Stern School of Business.

Kristen also recently published a book about data visualization, “Data Visualization Made Simple, Insights into becoming visual”

Kristen is an absolute expert on Data Visualization and teaches data viz best practices for both NYU students, as well as through a certificate program. Her passion is in helping up-and-coming analysts use visualization to enhance their work, tell stories, and communicate effectively with data.

Data visualization is important because we can use it to:

1) Explore our data and understand it

2) Communicate well, especially with non-data-literate people

In the former, when you’re exploring your data, you want to use more rudimentary visualization tools like scatterplots and trellis plots. These are great for understanding variation or differences between dimensions. But they are pretty terrible when it comes time to present your findings.

“Don’t make your audience work too hard”

— Kristen Sosulski

For the latter, when using data viz for communication, stick to simpler methods like bar charts, line graphs, and maps. Preattentive attributes, highlighting the thing you want someone to focus on, is a really effective way to keep someone’s attention.

So how about highly designed visualizations like Infographics? Kristen wouldn’t say “no”, but she certainly wasn’t wild about them. The problem is that they tend to over-simplify the data that it’s trying to communicate. That said, there are some great design concepts that we can use from infographics when creating powerpoints and other presentations.

While we cover some in-depth topics, it’s clear that data viz is for everyone, not just data science. Kristen covers the 4 categories of data viz tools, from basic excel or powerpoint, to advanced like R or Python.

We also wanted to learn more about Kristen’s new book, “Data Visualization Made Simple”. In chapter 6, we learned about ways to maximize retention of the reader. A critical piece to this is EMPATHY and being in-tune from your audience. You may go as far as drafting a survey so that you can understand the potential reader and make sure you’re designing to their needs. Don’t make your audience work too hard.

We wrapped up our conversation, talking about the future of visualization, and discussing how things like augmented reality and AI are already starting to change the game for data viz.

Thanks for coming on the show, Kristen!

More about Kristen

Check out her book on Amazon: Data Visualization Made Simple
Check out Kristen’s NYU Stern Class: Visualization Data
Follow Kristen on Twitter: @sosulski
Follow Kristen on LinkedIn: in/sosulski

Resources and Links from the Episode

Sankey Diagram
Trellis Plots
Understanding Media: The Extensions of Man by Marshall McLuhan
What I Talk About When I Talk About Running by Haruki Marukami
Amanda Cox from the New York Times’ “The Upshot” section
Science of Happiness Podcast
Data Science for Business by Foster Provost & Tom Fawcett
Edward Tufte
Nathan Yao
Stephen Few
Dona Wong

Learn more

Permalink

How to Data Viz like a Pro Part 2

Beyond the Data — Tue, 08 Jan 2019 13:30:00 +0000

Listen to the Episode

Subscribe to the Podcast

iTunes Spotify Stitcher Pkt Cast

Episode Summary

If a picture is worth a thousand words, then a data visualization must be worth far more than that.

In Part 1 of our two part series on data visualization, we talked about GOOD visualizations, what types of visualizations work better, when to use them and the like.

Today we’re talking about BAD visualization. When data viz goes wrong. And of course we have to start with the PIE CHART. As a wise friend once told me, “If a chart is named after food, then I don’t like it”.

If a picture is worth a thousand words, then a data visualization must be worth far more than that - Dave Mathias

We know lots of people don’t feel the same way about pie charts, so we wanted to discuss a bit about WHY it’s not a great tool for helping you tell your data stories. We won’t say you can’t use it, but make sure you know what it does and doesn’t do well. We’ll also hit on the “Data to Ink Ratio” which was pioneered by Edward Tufte and look at the pie chart on this ratio scale.

Finally, we wanted to talk about DESIGN when it comes to data visualization. Design doesn’t have to be colorful or frilly. Design can actually be minimalistic and utilitarian in form and function. The goal here isn’t to say that one is better than the other, but to ensure you’re thinking about your audience and how you want them to act after seeing your visualization.

If you’re creating something public and want lots of Shares, Re-Tweets and Likes, then a more infographic approach can work well. If you’re creating something for your CFO, tables, numbers and no-frill visualizations are probably a better way to go.

Resources and Links

Some great resources that can help you get started are Storytelling with Data by Cole Nussbaumer-Knaflic, and Makeover Monday by Andy Kriebel and Eva Murray.

We Deserve a Better Paradigm for Professional Education

Beyond the Data — Mon, 07 Jan 2019 13:00:00 +0000

We Deserve a New Paradigm For Professional Education

Providing new and innovative ways to deliver data training is one of the founding tenets of Beyond the Data

Higher education is in need of disruption. Decade after decade it remains essentially unchanged. An educator stands up in front of students and dictates knowledge. The students’ knowledge of facts, theories, and processes with occasional application are then tested.

Worse yet education has become increasingly expensive with students investing large sums prior to truly knowing what they want to do. Then, they go off into the workplace and in land of rapidly changing environments many times those skills become obsolete.

One of the founding tenets of Beyond the Data was to find a better way to provide the RIGHT skills to the RIGHT people at the RIGHT time. Starting today, we’re re-writing the rules on professional education

The building blocks of a new education paradigm

Affordable

If this is going to work, then it needs to be affordable for both students directly and also for employers paying for employees’ education. We’ve seen the mountain of debt that students come out of school with. It can’t continue like this.

Accessible

Students should have opportunity to learn no matter where they are in a convenient fashion. This means not having to drive long distances to stale classrooms. It can mean online classes, but it could also mean learn-at-your-own-pace type environments. Or more one-on-one scheduled mentoring sessions.

Practical

If a student can’t apply the knowledge in some meaningful way RIGHT NOW, then what’s the point? Providing real problems that they are passionate about is what will create lasting skills that improve their careers. It is time to stop memorizing facts and to stop thinking in theoreticals.

Continual

Learning doesn’t stop when you leave the classroom. In fact, it might not START when you enter the classroom. Learning takes time and requires doing, seeing, experiencing, and discussing. That’s why the lessons should be long-lasting, with the content always available to come back to… months or even years later.

Communal

This might be the most important part. Learning happens in a shared space with others. When communities are created, ideas are shared, relationships are built and we become better with these people than we ever could have without. They push us to think differently, to reach beyond our limits. Community is the secret sauce that makes learning work.

- Dave Mathias

Follow me on LinkedIn

contact me

How to Data Viz like a Pro Part 1

Beyond the Data — Thu, 03 Jan 2019 13:30:00 +0000

How to Data Viz like a Pro Part 1

Episode 009

If a picture is worth a thousand words, then a data visualization must be worth far more than that - Dave Mathias

If a picture is worth a thousand words, then a data visualization must be worth far more than that.

People respond to pictures. There’s an emotional reaction that drives action and decisions. Black & White numbers on a page will only take you so far in making your users actually do something!

In Episode 1 of this two part series, we talk about the the two core ways that dataviz can help. The first is Dataviz for exploring your data. Using tools like scatterplots and small multiples will help you find the outliers… to FIND the story that needs to be told in your data. But these don’t do a great job of quickly and easily telling your story. You have to search for the answer.

That’s why there’s a second type of dataviz that we want to discuss… specific for telling a compelling story. Maps, bar charts, and line charts are going to be your bread and butter here.

We also talk about the importance of communicating precision, confidence, or error bands and the various ways that you can help the reader understand how accurate your data might be.

For inspiration, we encourage you to check out Storytelling with Data by Cole Nussbaumer-Knaflic, and Makeover Monday by Andy Kriebel and Eva Murray. They have tons of great content on how to think about (and practice!) good data visualization.

Next week’s episode, we’ll talk about some of the risks in using visualization to easily mislead or lie to your data consumer.

Thanks and Happy Listening!

Subscribe to the Podcast

Permalink

Beyond the Data Attends MinneFRAMA 2018

Beyond the Data — Thu, 06 Dec 2018 08:03:07 +0000

Today, we’re discussing Dave and Matt’s experience at MinneFRAMA 2018, hosted by the always wonderful, MinneAnalytics.

Dave:

So Matt, we were fortunate enough to attend and actively participate in MinneFRAMA 2018 this year. The event was geared towards financial, retail, and marketing analytics, and people across the Twin Cities showed up in droves!

Matt:

Yes! This was the first event that MinneAnalytics has hosted in St. Paul and it did not disappoint. The Science Museum of MN hosted us and it was a great location.

Dave:

There’s just something about walking into an analytics conference and being greeted by a life-size T-Rex that gets you in the mood for some data crunching.

Matt:

I’m not sure it made me want to crunch numbers, but it definitely made me want to record an episode of the Data Able podcast! What an amazing view we had while we sipped our morning coffee and discussed the finer points of using data effectively.

Dave:

We were lucky enough to have Tessa Enns and Liz Weber join our show, live. They were such gracious guests. I wish we could have talked longer with them. We should probably move along with our Top 10 list from the event, huh?

Matt:

Yes we should. Since you spoke at three different sessions, why don’t you relax a bit and let me run with this one. Without further ado…

Beyond the Data’s Top 10 List from MinneFRAMA 2018

The Science Museum of Minnesota was a hit.
Tons of great conversations that were helped by the fantastic space. We made music all day by ascending and descending the musical stairs. Plus, it was a huge bonus to get a free ticket to a future Science Museum visit!
AI was all the buzz.
The hype is strong with AI at MinneFRAMA this year. However, the applications went deep, including discussions from the future of work, augmenting attorneys, and even AI to monitor AI.
The Future of data privacy regulation is uncertain, but direction is not.
With Europe's General Data Protection Regulation (GDPR) and the California Consumer Privacy Act in place, the general view is this only the start. Good discussion around data privacy and its present and future with Melissa Krasnow moderating a panel of experts, Tim Nagle, Erich Axmacher, and Brad Hammer.
Data storytelling and viz is in high demand.
The importance of data visualization and storytelling was a popular theme. Arlene Birt's Visual Storytelling: Putting Data into Context session was a hit and great turnout for the TC Data Viz MeetUp with an engaged audience right after lunch.
Not everyone needs a screen and slides.
This is the first time where MinneAnalytics had a room that was designed to not have a screen! Lots of great sessions were held in an informal setting at the Elements Café, overlooking the Mississippi river. Great job by Josh Moe and Morgan Catlin using a whiteboard to tell their story. We also saw post-it notes and pinned print outs!
Sharing a meal with She Talks Data.
The wonderful She Talks Data MeetUp held a networking session over lunch. Packed from the start, it was a popular destination. Thank you to Serena Roberts and Laura Madsen for continuing to advocate for women in data!
Startups abound.
We had more startups participate in the MinneFRAMA Startup Showcase than any of our prior events. Analytics knows no company size boundaries and is often used by startups as their disruptor.
Getting technical.
While Data Tech is MinneAnalytics most technical event, MinneFRAMA had many great technical sessions starting off with sessions like Joe Konstan, PhD and ending with Jason McNellis. Like all MinneAnalytics events the goal is to provide a variety of options.
Standing room only.
While I can't speak to all the sessions, Jason Rogowski and Ryan Stellmaker's session around Building Marketing Analytics Capabilities, Brick by Brick was the likely winner for biggest audience - standing room only in the Omni Theater. Impressive Jason and Ryan!
Live podcast taping to kick things off.
Last year we had Kyle Polich record an interview with Joe Konstan, PhD on Data Skeptic at FARCON, this year we got to interview Liz Weber and Tessa Enns on the Data Able podcast. Liz and Tessa told some great stories about successful analytics projects. Make sure to subscribe to Data Able on your favorite podcast catcher and listen to the MinneFRAMA taping when it’s released.

Listen to our Podcast

So what is our big takeaway from all of this? It’s all about community!

While sessions were often full, there were many hallway conversations both between and during sessions. Tons of engagement with people making new connections and re-invigorating old connections. There was even a mini-job fair where recruiters were talking with interested persons. And of course, the day ended with good beer, good wine, and good conversations.

All in all, we had another stellar MinneAnalytics event thanks to the presenters, sponsors, and most importantly the attendees.

Until next year!

Creativity plus Analytics equals Amazing

Beyond the Data — Tue, 27 Nov 2018 13:00:00 +0000

Creativity plus Analytics equals Amazing

Episode 008

Creativity and analytics. Think that's like oil and water? More like Peanut Butter and Jelly!

Often times, data and analytics are thought of as black and white concepts. They are the OPPOSITE of creative thinking. They provide facts, reason, and cut through the emotion.

But thinking about it that way really limits what data can do.

In this episode, Dave and Matt will talk about curiosity, creativity and the type of people that will be most successful in an analytics role. Hint: it’s the people that can think outside the box, come up with new ways to solve problems, and creatively communicate what’s going on in their organization

We also touch on the concept of “Design-Focus” and “Human-Centered Design”. It’s the belief that everyone has creativity in them, and with basic tools, can create new and wonderful things. It’s making sure that whatever is created has the end-user in mind.

It’s also important to balance colorful & whimsical design vs. modern and functional design. When creating something like a dashboard, it’s important to ensure it’s both functional to read AND pleasurable to view. Design without purpose isn’t useful.

Until next week!

Thanks and Happy Listening!

Subscribe to the Podcast

Permalink

Improve your analytics skills with... Improv?

Beyond the Data — Tue, 20 Nov 2018 14:00:00 +0000

Improve your analytics skills with... Improv?

Episode 007

Improv: The ability to think on your feet, ask good questions, and connect with your audience in a meaningful way.

Everyone wants to get better at their jobs, right? For people who use data in their work, they usually look to more technical training… Excel, Tableau, Python, R, SQL, Qlik, PowerBI, etc.

What if I told you that you could get the same, if not MORE, useful training by taking an Improv course at your local comedy troupe?

Yep, we said it. And we stand by it too. Improv helps you think on your feet, ask good questions, and connect with your audience in meaningful ways. Does that sound like something useful to an analyst?

If you’re trying to bring data to your organization, then being able to captivate your audience, communicate effectively with the person across from you, and respond appropriately and confidently goes a LONG way towards moving the needle.

Perhaps you’re asked to present your analysis to some executives at your organization. You’ve worked tirelessly on the deck and your speaking points. And within 5 minutes, the execs have asked a ton of questions, thrown off your script, and moved the conversation in an unexpected direction. Wouldn’t it be nice to have the skills to react quickly, keep your composure, and respond to their new and changing topics?

That’s Improv, baby! Go find your local comedy club and sign up for a course today.

Until next week!

Thanks and Happy Listening!

Subscribe to the Podcast

Permalink

The making of a good analytics leader pt 2

Beyond the Data — Tue, 13 Nov 2018 14:00:00 +0000

The making of a good analytics leader pt 2

Episode 006

What is the makeup of a good analytics leader? How do leaders become successful in leading analytics teams?

Make sure you check out Good Analytics Leader Part 1 that focuses on executive leaders.

What does it take to be a great front-line leader for a team of analysts?

Leading a team of analysts is a rewarding but can be challenging as well. Many analysts are great with numbers, math, code, and visualizations, but can sometimes lack the softer skills like effectively communication, project management, or requirements gathering.

These are necessary skills, and you as their leader must help them get there! But also have empathy. Recognize each person’s individual strengths and opportunities and then position them to leverage their strengths and minimize the opportunities. Recognize that your amazing SQL developer shouldn’t (and probably doesn’t want to) be put in charge of project managing your biggest deliverable.

There are also some critical skills that you might need for yourself. First, an ability to change quickly. The analytics field is shifting very quickly. New methods, new tools, your team being hired away by bigger companies with seemingly endless pocket books.

You’ll also need to be VERY good at being a champion, spokesperson, and advocate for your team, and for the work they do. Most front-line teams (and even executives) don’t really get what your team does. It’s up to you to sing from the rooftops all the ways that your team adds value.

Managing is a journey, and you won’t be good at it on day 1, but as long as you are channeling your team’s successes and put them in the best position possible, you’ll do great.

Until next week!

Thanks and Happy Listening!

Subscribe to the Podcast

Permalink

The making of a good analytics leader pt 1

Beyond the Data — Tue, 06 Nov 2018 14:00:00 +0000

The making of a good analytics leader pt 1

Episode 005

What is the makeup of a good analytics leader? How do leaders become successful in leading organizational analytics?

Whether you’re in analytics, product, finance, operations, or any other department, there are lots of good leaders out there. There are also lots of bad ones.

In this two-part series, we’ll be exploring the analytics leader, how it might be the same or different from other types of leaders in other departments, and what it looks like.

In this week’s episode, we’re going to focus on the top-level leader. The “Chief Data Officer” if you will. Whether they actually have the title, or something like “VP of Analytics”, or “Director of Customer Insights”, someone at your organization is playing the CDO-role whether you know it or not.

So where do these types of people come from? How do you become a Chief Data Officer? Do they come up the ranks of the analyst track? Or do they come from other disciplines, and they just happen to understand data as well?

Regardless of where they come from, the most important thing that a CDO-role will need to do is make sure that the organization as a whole is thinking “data-first”. This means consistently challenging the gut decisions of her C-Suite peers. The CFO might state that “We know that our customers want cheaper prices”. Do they? What data led us to this conclusion? Can the analytics leader help bring data to the table to verify?

The great leader knows how to place themselves in the right conversations, and then make sure that data is a part of that conversation.

It’s about building the CULTURE of analytics. It starts at the top, with executives, but it also means they need to lead a capable data team, and ensure that each line of business is being served and that they are capable of doing something with the data once they have it.

It’s not an easy position to be in, but it’s certain a necessary one for any organization who wants to be more data-informed.

Until next week!

Thanks and Happy Listening!

Subscribe to the Podcast

Permalink

The data driven and data informed culture

Beyond the Data — Tue, 30 Oct 2018 13:00:00 +0000

The data driven and data informed culture

Episode 004

What does a data culture mean? And is there a difference between Data-driven vs. Data-informed?

An organization that uses data in their decision-making process. From the highest executive to the summer intern.

Sound far-fetched? It shouldn’t! Most organizations are on their own journey towards using data to drive significant value for their customers and shareholders.

But many organizations aren’t there yet. And that can be frustrating. Building a data-informed culture doesn’t happen overnight, but it does lead to great results.

In this episode of Data Able, we talk about the Data-driven culture, and the Data-informed culture. Mostly semantics, the difference is in the level of maturity the organization has with using data. Ideally, every business line is comfortable with combining their deep industry expertise with their data and insights that lead to the best decisions and outcomes possible.

Until next week!

Thanks and Happy Listening!

Subscribe to the Podcast

Permalink

The (un)ethics of using data in your organization

Beyond the Data — Thu, 25 Oct 2018 16:05:00 +0000

Looking back at history, we appreciate that just because something is legal, it does not mean it is ethical, and vice versa. For example, from recent history, Facebook may have satisfied legal requirements with Cambridge Analytica, but I think most of us would say what Facebook did was unethical.

Of course, decisions about data legal versus data ethical are not made in a vacuum. There are many changing factors and competing interests that blur the line. Some of these factors and interests are the quickly changing dichotomy of global views on data privacy rights, changing demographics and views of transparency and privacy, information and physical security, and quickly changing legal and regulatory environments. These items lead to significant questions businesses and governments are facing around data and how to use it.

This leaves a lot of questions.

How should organizations react?
Does it matter if these actions are for our own “good”?
What if these actions are for people's or society’s own “good”?
Does it matter if governments or organizations are transparent?
Can the ends justify the means?

The answers to these questions are not always so black and white. However, some organizations like Microsoft are taking the long view, in my opinion, and encouraging the world to align with General Data Protection Regulation (GDPR) standards. While others strongly oppose changes in line with the protectionist data privacy rights that GDPR provides.

I hope that society and organizations will be radically transparent in how data is collected and used. This means communicating with people simply and clearly. Doing this is not only good for people but also good for companies and society at large. Consumers are shifting and want ethical, transparent, and trustworthy organizations to support them. This is especially true for Millennials and Gen Z.

If organizations are radically transparent and simple on how data is being collected and used, then it is up to us as citizens to take the next step. Stop supporting companies, people, and parties that don't align with our beliefs. Doing so will result in beneficial change. At the same time, make sure not to get caught up in reacting to news stories without knowing the facts, as too often sensationalized stories sell over boring explanations.

Now, it is time for you to make a difference and not accept data legally to trump data ethics. If you are an individual contributor and believe your organization’s actions are unethical - leave! If you lead a team or organization that handles data - be transparent and communicate simply! If you are a manager with a team member who does not respect the data in your stewardship - educate and transform him! If you are a politician with a bill coming before you that seeks to remove data transparency, vote no!

Data legal versus data ethics is not just a matter of opinion or of profit. It is a matter of value and culture. Which way will you steer your ship?

- Dave Mathias

Follow me on LinkedIn

Need help?

contact me

The exciting future of self service analytics

Beyond the Data — Tue, 23 Oct 2018 12:00:00 +0000

The exciting future of self service analytics

Episode 003

The future of self-service analytics is bright, but could things like AI help analysts do their jobs even better?

Self service analytics isn’t going away anytime soon. In fact, we think it’s only become more prevalent!

Why is this? Well for starters, there’s more data than ever before! And our BI, Analytics and Data Science teams just simply can’t (and shouldn’t) keep up with the demand. This is a great problem to have, but will require shifts in the traditional analyst paradigm.

Leaders are finally starting see the value in their data, and in order to get them what they need, we need to move faster, getting the RIGHT data, in the RIGHT hands, at the RIGHT time.

So what about Artificial Intelligence? Is it going to eliminate the need for analysts, data science, etc.? The short answer is no… but it IS going to require that the consumers of this information are capable of a baseline understanding of how data works, how math works, and how to communicate what’s being created.

We hope you enjoy this episode. Until next week!

Thanks and Happy Listening!

Subscribe to the Podcast

Permalink

Twin Cities Meetups Are Creating a Data Culture

Beyond the Data — Thu, 18 Oct 2018 16:05:00 +0000

This month the Twin Cities hosted Startup Week. One of the many great sessions was about the history of data visualization, delivered by Matt Dubay on behalf of the Twin Cities Data Visualization Group (TC Data Viz) (of which I’m an organizer). There was an amazing level of engagement and energy in the room and for the topic. As a side-note, I encourage you to check out the TC Data Viz group if you haven’t already. It’s a place where all are welcome - beginner or advanced - technical or business users – and our tools are agnostic, whether open source or proprietary software. We provide a fun and creative space to share ways to display data in our businesses and communities.

One of the main themes discussed during the session was around data literacy. One astute person noted that IT and self-service BI cannot drive the destiny of data literacy within organizations. Instead, it needs to be something solved by the business users themselves. Only then will self-service BI truly succeed. I couldn't agree more! Data fluency is an upcoming challenge that business leaders, managers, and individuals need to make time for, if they’re going to create data-savvy organizations.

In related news, the September 2018 McKinsey Quarterly published an article entitled "Why data culture matters". The whole article is great and highly recommended, but one key insight I want to touch on was that “Data culture is decision culture". The takeaway here is that organizations shouldn’t "… approach data analysis as a cool 'science experiment' or an exercise in amassing data for data's sake. The fundamental objective in collecting, analyzing, and deploying data is to make better decisions". One other thing I want to touch on is their call for the “democratization of data" and its importance in a data culture. From the article: “… get data in front of people and they get excited. But building cool experiments or imposing tools top-down doesn't cut it. To create a competitive advantage, stimulate demand for data from the grass roots."

Certainly, executive buy-in is important for resource allocation and overarching strategy, but executives don't make most decisions. Organizations succeed by the many decisions each employee, contractor, and customer make each day. Empowering and encouraging those stakeholders to get excited about data whether it is educational opportunities, competitions, data-for-good initiatives, or other ways to help invigorate and empower data culture at the grassroots level is essential.

So a little homework this week:

Executive: Identify a way to empower and encourage your organization to support a grassroots-level data culture. What change can you support and encourage at the grassroots level so that everyone not only wants data but needs data to survive?

Managers: Identify a way you can empower and encourage your team to support a grassroots-level data culture. What new decisions can you or your team harness new or existing data to make better decisions than you had before?

Experienced contributors: Identify how you can better use data to make better decisions and even demand data that had not been used before to make decisions? Further, how you can you help support newer contributors in this effort?

New contributors: Provide a fresh insight on how your organization can better encourage using data in roles. Your fresh perspective has a distinct advantage of seeing what could be as your are not encumbered by what is or was.

Now go and do your part changing the data culture at your organization!

BEYOND THE DATA IS ON A MISSION

We help high-performing individuals become champions for a more data-driven approach in their organization. We believe that data science is only part of the equation.

Getting value out of data requires building a culture that starts with YOU, is supported by executives, and trickles down to every front-line specialist in your organization.

Is it time for everyone to be data literate?

Beyond the Data — Tue, 16 Oct 2018 12:00:00 +0000

Is it time for everyone to be data literate?

Episode 002

Data literacy is a fast-growing topic. But what is it? And why should organizations care?

So what is this new concept we’re hearing a lot about these days? Data Fluency or Data Literacy is getting a lot of buzz right now. What is it and why should i care?

We describe it as: Framing problems, applying data, making data-informed decisions, and being able to communicate with data (through storytelling, or data visualization).

Another key aspect of a good data user that isn’t taught in textbooks or online coding classes? Empathy. The ability of the data user to understand the needs of her audience, craft the right narrative and deliver the right answer to the right person at the right time. You may even consider putting empathy in your job descriptions going forward!

So is data literacy only for analysts and data scientists? Absolutely not! We think product teams can benefit from using data. We think HR teams, Finance teams, Operations teams, Sales teams and Marketing teams can all benefit from data literacy. It’s a tool in your toolbelt to help you become a better marketer, finance leader, or product owner.

We would be remiss if we didn’t share a few of our favorite resources for further study. First, there’s Cole Nussbaumer’s fantastic blog Storytelling with Data. We also love Kate Strachnyi’s Story by Data blog. And check out Jane Crofts’ company, Data to the People who are building data literacy assessments for organizations all over the world!

Thanks and Happy Listening!

Subscribe to the Podcast

Permalink

At the Intersection of Data Literacy & Design Thinking

Beyond the Data — Thu, 11 Oct 2018 14:40:00 +0000

Design thinking is an approach made famous by IDEO and Stanford’s d.school. The premise is everyone is creative and that the human should be at the center of design. It aims to provide a framework to design things that are desirable, feasible, and viable. The design thinking approach is a six-step process of framing a question, gathering inspiration, generating ideas, making ideas tangible, testing to learn, and sharing the story. Going through this process can be linear but often isn’t.

Now ask are data fluency and design thinking similar? The answer is definitively yes! Design thinking is a mentality of framing problems, generating ideas, testing ideas and telling stories. Data fluency stresses a similar process. Further, when determining desirable, feasible, and viable you certainly need to understand what the data indicates. Additionally, data fluency is always focused on the pain of people whether internal or external customers just like design thinking has the human in the center and their pain and needs.

The other thing design thinking and data fluency have in common is they are both geared at democratizing out their practices to everyone. Design thinking aims to put design in the hands of everyone while data fluency aims to put data science in the hands of everyone. This of course brings about fear by some in respective professions but it really brings about opportunity for all. This practice democratization solidifies importance and adoption. There will always be a place for those specially skilled in the respective design and data science arts, but it is time for basic practices and understandings of both to be adopted by all.

If you are a data person but looking to learn about design thinking and human-centered design pioneered at IDEO. Or, if up to the challenge take the 90-minute virtual design thinking crash course at d.school. In a future post we will go into how design thinking practices can be used in data fluency when we discuss the Data Value Cycle.

What is this self-service BI thing?

Beyond the Data — Tue, 09 Oct 2018 12:00:00 +0000

What is this self-service BI thing?

Episode 001

What in the world is Self-service BI and why should anyone care about it? Is it just for analyst-types?

One of our favorite topics to kick off this podcast! Matt has been building self-service analytics initiatives in Fortune 500 companies for 7 years now, so he’s excited to talk about it with you!

The foundation of self-service analytics is in allowing non-technical users throughout an organization to access and use data to identify insights, communicate results, and drive decision making.

Why would an organization choose to go this route? What are all those analysts getting paid to do, anyways? Well, in the age of information, there’s far more data than any single analyst team will be able to mine, process, and communicate. You want your technical people working on the really challenging problems like sourcing new data, cleaning data, and more advanced statistics & machine learning.

What this does is it frees up the business to move at THEIR speed. If they need an answer now, then you’ve given them the tools to get that answer, rather than being put in a queue. You want the data as close to the decision-making as possible, and as fast as you can. Democratizing the data can help you make that happen.

Make sure to subscribe to more episodes with your favorite podcast catcher!

Thanks and Happy Listening!

Subscribe to the Podcast

Permalink

Name it and they will come

Beyond the Data — Tue, 25 Sep 2018 12:00:00 +0000

Name it and they will come

Episode 000

We are on a mission to help high-performing individuals become champions for a more data-driven approach in their organizations

Welcome to the Data Able Podcast!

Dave Mathias and Matt Jesser are proud to bring you a brand new podcast all about data. our goal is to help people like you to champion data in your organization and truly transform yourself into a data-informed culture.

How are we going to do this?

We think podcasts are a great way to provide high-quality information in a tight and compact way. So we’re going to be delivering weekly, quick-hit episodes on topics that your organization needs to be thinking about. These 10-minute episodes are easy listens, and will arm you with ideas and talking-points that you can use to drive the data-culture in your organization.

We’ll also deliver longer-form interviews with amazing people with a similar passion for data, who are doing work just like you… helping change your organization with data.

Our first (zero-eth?) episode is a light-hearted look at how we came up with the name for our podcast, and what we’re looking to achieve.

We look forward to starting this podcasting journey with you! Until next week!

Thanks and Happy Listening!

Subscribe to the Podcast

Permalink

Why it's critical that leaders become data literate

Beyond the Data — Thu, 09 Aug 2018 18:09:00 +0000

Nearly every organization is undergoing a digital transformation and as part of this data literacy or data fluency plays a pivotal role. Data like any language is effective when others around you understand it and make decisions based on its meaning. But, data fluent doesn't mean being a data scientist. Instead it means "the ability to understand and use data effectively to inform decisions" according to Mandinach and Gummer. [1] One addition to this definition would be ability to communicate with data.

Leaders with data fluency whether team leaders, department directors, or senior executives benefit. These data fluent leaders ask questions like those below but more importantly are able to make data informed decisions.

What are key metrics that help me understand my customer's experience?
Am I hiring, rewarding, promoting and training my team members to be data fluent?
What data can I share with others to empower them to make the organization better?
Am I being a good data steward and ensuring proper data privacy and ethics are being utilized?
How can I use data to make our operations more efficient and effective?
Am I communicating with data appropriately to show the value our organization
What new data could I seek out or capture to bring more organizational value?
What percentage of employees have access to self service business intelligence and analytics and have been trained on it?
What metrics do we track to measure our employee experience?
What percentage of data we capture are we using to inform decisions?
I understand my NPS is in the top quartile, but what is driving this metric and what other metrics should I be monitoring to understand my customer satisfaction?
How are we developing new products and services based upon data from our customers?

Data fluent leaders are able to help their organizations have a data driven or data informed culture. Doing so will not only lead to more fulfilling environment and to great success.

Are you a leader interested in helping your organization be data fluent? Reach out and let’s discuss if we can help.

[1] McAuley, D., Rahemtulla, H., Goulding, J., & Souch, C. (2014). How Open Data, data literacy and Linked Data will revolutionise higher education. Retrieved from: http://pearsonblueskies.com/ 2011/how-open-data-data-literacy-and-linked-data-will-revolutionise-higher-education/

How business and tech partners can better work together

Beyond the Data — Thu, 24 May 2018 20:26:00 +0000

The Twin Cities Data Fluency Group had its second meeting in May. This month involved an engaging discussion on “How the business can better work with analytics and tech partners.” Tricia Duncan and myself (Dave Mathias) moderated three great panelists – Nate Hallquist from Syngenta, Serena Roberts from Capella, and Jack Vishneski from ThreeBridge and consulting with Cargill.

There was a lively discussion on several fronts, but key takeaways were as follows:

Building relationships is key. Most information work takes teams and that means working with people. The more you build relationships the better chance to succeed as Nate mentioned.
Bring everything back to problem being solved. Data and analytics only serve a purpose if they solve problems. As Jack succinctly mentioned it is all about solving problems and bringing conversations back to those problems will help ensure success.
Trust is key. As Serena mentioned being a trusted advisor as an analyst and business partner alike is a must. Serena has the unique experience playing both roles in sales and sales enablement and building trust with both these hats has been essential to her success.
Rapid prototyping should be norm. Rapid prototyping is a must for dashboards and both to help ensure customer satisfaction and efficiency. These rapid prototypes can be done in a dashboard tool if a similar dataset available but just as nice it can be hand drawn on a whiteboard or paper.

In addition to these takeaways, there was a good discussion on the role of self-service business intelligence (BI) and how much autonomy the business should have and how much of it stays in the analyst, data science, or technology hands. There was mixed feeling here both on panel and in audience. Some companies have shown more success than others in distributing data fluency and technology into the business. However, there was agreement that tools are making it more able for end users to do more challenging problems.

One metaphor that seemed to resonate is treating self-service BI as a grocery store and not a treasure chest can help. As Nate described this the analyst, technology, or data science groups ensure that often used data has been made available with appropriate cleaning, integrity, and trust to business users. However, organizations need to ensure end users have proper training, tools, and help available so they can focus on conversations and insights while reducing the risk of invalid data models or technical debt.

There was a lot of overall agreement that data fluency is critical for organizations broadly and the language of data will be more easily picked up by some than others. But, to have a data-driven or data-informed culture at an organization requires your people to be data fluent.

This is a short summary of the great discussion that occurred, and all are welcome to attend the next TC Data Fluency MeetUp will be in July (date TBD). If you are an analyst or data scientist, then this is a great opportunity to bring one or more of your business partners to help further your relationship.

Thank you to Nate, Serena, Jack, and everyone that attended, and Tricia, Nate, and I hope to see you in July.

She Talks Data Perspective From a He

Beyond the Data — Thu, 22 Feb 2018 20:17:00 +0000

Earlier this month I had the opportunity to attend the She Talks Data MeetUp in the Twin Cities. This group’s goal is “[o]ur goal is to build a close-knit community of women (and men in support of women) who can come together to grow professionally and personally.” It was started as an offshoot of the She Talks Data group in Silicon Valley a few months ago locally by Serena Roberts and Laura Madsen.

Serena has said several times that this is not just for women. Plus, one of my friends, Karla Hillier, was presenting, so I thought great to support her and at same time attend this new group and learn.

As I told Serena beforehand, I was afraid attending and how I would be received and feel. But my fear quickly dissipated from the moment I walked in. Right from the start it was an engaging and welcoming environment, but I did feel something different.

As the first speaker Emma Denny, an employment law attorney, kicked off right from the start the room was riveted. There were questions related to workplace discrimination and sexual harassment. Emma talked about rights that people had in Minnesota. But, she also talked about the high thresholds that people face in these cases and difficulty in proving these cases. There were great tips such as telling people in writing when they felt harassed and literally spelling it out that you think it is because of gender or other protected class.

At one point, Emma asked how many people in the room had felt discriminated or harassed at work and nearly everyone’s hand was raised. I can say I felt bummed and really more angry. I felt angry that so many talented amazing people in our community have felt discrimination and harassment. I felt angry that so many of amazing people will likely face this more as their career continues. I felt angry that people often time creating those environments are oblivious that it is even occurring until it is too late or worse don’t care.

After the group I reflected what I could do. Yes, as a product person at heart my nature is when I see a problem I want to help find a solution. Of course, there is no single solution, but we can all help one action at a time whether in groups or at work to provide a more inclusive environment.

I encourage other men to respectfully participate in She Talks Data and other groups like these where appropriate and where welcomed. Not only as a sign of support, but also to be in a better position ourselves to be supportive when challenging situations with bosses, colleagues, employees, and clients will inevitably occur. After all we are people on this journey of life together with a finite amount of time, so let’s make the most of it and support each other through it.

Shout out to all the great people I met and good conversations I had. Special shout out to: Jen Roberts and Tricia Duncan that I had pleasure of meeting and sitting with; Serena Roberts and Laura Madsen for organizing group locally and continued leadership in community; and Emma Denny and Karla Hillier for sharing their knowledge and inspiring others.

Interested in learning more? Go to the April 4th She Talk Data MeetUp and catch April Seifert who will be one of the presenters. In fact, April and I were just talking this morning on all things CX, analytics, and podcasting and sure she will have a lot of great wisdom to share.

Data Informed Organizations Need Data Literate People

Beyond the Data — Thu, 18 Jan 2018 21:45:00 +0000

Per my discussion on Stages of the Data Transformation Journey from data-ignorant to data-informed, it is not a “one-size-fits-all” and is ongoing. However, it is a misconception that a data-informed organization only means having executive buy-in backed by an awesome data science team. Yes, both are important. But, you can be a data-informed organization without the most advanced data science team, or even without a data science team at all.

Instead, data-informed organizations require data-fluent people making decisions, and an empowered, data-informed culture behind it. Data-fluent people meaning people that can apply data, gain insights from data, and tell stories with and around data. Data informed and empowered culture meaning an organization that encourages and empowers data-informed decisions.

For example, a data-fluent project manager may seek out new data he typically does not receive to better ensure his project meets targets. Or a data-fluent product manager may utilize customer data insights to develop the next "big thing" for her organization.

Most decisions are not made by people with VP or CXO in their title. Instead, decisions are made by analysts, associates, engineers, customer service managers, generalists, architects, scientists, and dozens of other titles. It is the culmination of thousands of decisions each day that is the fate of organizations. Ideally, each of these decisions is made by a data-fluent person or persons in the context of customer value. (Look for more to come on the subject of being customer-focused).

“Most decisions are not made by people with VP or CXO titles”

The real value of data-fluent people, though, is not what they do alone. Instead, data-fluent people help drive improvements for the entire organization. And that makes the organization run better. Think what companies like Toyota did on the factory floor — but now it is on the office floor.

In addition to encouraging specific improvements to processes and systems, data-fluent people help drive a data-fluent culture. This is where data-fluent people provide the highest value.

The important question then is, does your organization hire for, train towards, and reward data fluency? If the answer is no, then your organization will not be data informed.

So, the next time an executive asks how do we become a data-informed organization? Respond, “We need to start attracting, retaining, developing, and empowering data-fluent people."

Stages of the Data Transformation Journey

Beyond the Data — Mon, 15 Jan 2018 15:15:00 +0000

The data transformation journey moves companies from data-ignorant to data-reactive to data-driven to data-informed and, in fact, eventually to real-time data-informed. The data transformation journey is a process. This journey isn’t straight; it will take some organizations longer than others to progress. Organizations also generally don’t move in lockstep in this journey. But to be clear, companies that don’t progress along this journey will eventually not exist.

That is a lot for one paragraph, so what are these data maturity stages:

Data Ignorant: Data ignorance means your organization doesn’t meaningfully use data. The organization may be getting data but doesn’t understand quality, meaning, and/or context. The organization may have great people, but their true north is their gut or their manager or executives' gut. There are still many successful companies doing this. Leaders with experience and intuition can often add enough value to overcome the influence of data for some time. Just remember that, yes, ignorance may seem bliss, but data ignorance is only bliss until you have competitors that are not.

Data Reactive: Data reactive is just what it sounds like - organizations capture data but don’t strategically use data in their decision-making; instead, they use data to react. Further, they don’t strategically define the data to capture but rely on what others do and copy it. For example, a T-shirt company realizes it lost 34% market share in a market it had once dominated. Now they react and make a salesperson change and a social media ad buy. Data-reactive organizations or departments are still very prevalent. They often have founders or leaders that rely heavily on gut still. They have also realized that complete data ignorance can bring, and they have moved forward to data reactive.

Data Driven: Data-driven organizations understand that data is valuable and almost always good data is better than any one or several people’s experiences. Data-driven organizations understand that they must be thoughtful in capturing data and then have processes in place where people use this data in their jobs to make decisions. Many data-driven organizations are extremely data-sophisticated by most standards. The challenge with data-driven organizations is that they often let the data drive them without a full understanding of the context of the data.

Data Informed: Data-informed means data is thoughtfully understood and processes aligned to maximize data-informed decisions in an organization. Further, data captured and other data sought are strategically determined to add organization and customer value. Data-informed organizations not only make decisions based on data, but they understand the context of data in those decisions. This context is provided from a combination of organizational experience, competitive landscape, industry expertise, and decision impact on various stakeholders.

Moving along this data transformation journey is not simple and is not one-size-fits-all. It is not something you can just ask how much it will cost to get me to the end of this journey. This journey is constant and doesn't have an end. Needs and technologies continue to change around data. Things like artificial intelligence, blockchain, information security, and quantum computing are currently and will continue to change the data transformation space. Further and most difficult, data transformation is a culture change for organizations.

Look for future postings where we will dive into further discussion on how to move along this data maturity journey. For now, I wish you Godspeed along this data transformation journey, no matter where you are.

Watch out for data science shiny objects

Beyond the Data — Thu, 07 Dec 2017 14:23:43 +0000

There are three main ways that organizations use data and analytics in their organization:

Enhancing customer/product experience
Enhancing employee experience
Increasing operational efficiencies

Individual data science efforts will often cross more than one of these areas.

I am writing this post because I see organizations making decisions more often based on increasing operational efficiencies than enhancing customer/product or employee experience. Cutting costs is the shiny object that never goes wrong on an earnings call. But does it add to the long-term value of your product, your organization, and your most valuable assets, your employees? Often, the answer is a short-term yes but a long-term no.

Be a leader that delivers customer and employee value from the power of data and analytics first. Along the way, some operational efficiencies will come along as part of your efforts.

Thoughts

Twelve Great Non-’Product’ and Non-’Data’ Podcasts for Product, Design, and Data People

- Dave Mathias

Need help?

Unlocking the Potential of Generative AI: 10 Prompt Engineering Recommendations

- Dave Mathias

Need help?

Integrating Generative AI: Questions to Answer for an Existing Product

- Dave Mathias

Need help?

Harnessing the Power of AI in Product Development

- Dave Mathias

Need help?

Product Roadmap Success Practices

- Dave Mathias

Need help?

My New Book! Data Storytelling and Translation: Bridging the Gap from Numbers to Narratives is available!

- Dave Mathias

Get your copy of Data Story and Translation today

We can all benefit from a little empathy and a map

- Dave Mathias

Need help?

Ep 44 - Haydee Hernandez - Building a culture of experimentation

Listen to the Episode

Subscribe to the Podcast

Episode Summary

Don't get bit by the Cobra Effect

- Dave Mathias

Need help?

Harness the power of surveys done well

Survey Design

Survey Execution:

Need help?

Do you believe in product magicians and data unicorns?

- Dave Mathias

Need help?

Ep 43 - Michel Guillet - Leveraging product and data-driven thinking to enhance sales teams

Listen to the Episode

Subscribe to the Podcast

Episode Summary

go Beyond the Data podcast

Ep 42 - Jansen Sullivan and Victor Anjos - Data education deserves better

Listen to the Episode

Subscribe to the Podcast

Episode Summary

Ep 41 - Jeff Richardson - Data strategy and leadership from a B2B CDO

Listen to the Episode

Subscribe to the Podcast

Episode Summary

It is time to put the BS in your organization, behavioral science that is

Need help?

Ep 40 - Jasmine RuKim - Creativity Deserves a Better Track Record

Listen to the Episode

Subscribe to the Podcast

Episode Summary

More about Jasmine

Ep 39 - Brent Dykes - Data Storytelling is More Than Data Visualizatio

Listen to the Episode

Subscribe to the Podcast

Episode Summary

More about Brent

Ep 38 - Allen Hillery - The Prolific Data Viz Writer

Listen to the Episode

Subscribe to the Podcast

Episode Summary

More about Allen Hillery

Transforming Your Data into a Compelling Story

Visual Data

About the Author

Allen Hillery

Adjunct Professor at Columbia University,Writer and Editor at Nightingale, a Medium.com Publication

Follow Allen:

Harnessing systems thinking to better understand and impact change

Need help?

Ep 37 - Rosa Pantoja - Evolution from a Data Champion of One

Listen to the Episode

Subscribe to the Podcast

Episode Summary

More about Rosa Pantoja

Data Viz Solves a Contradiction

Adjunct Professor at Columbia University,
Writer and Editor at Nightingale, a Medium.com Publication

ADJUNCT PROFESSOR AT COLUMBIA UNIVERSITY,
WRITER AND EDITOR AT NIGHTINGALE, A MEDIUM.COM PUBLICATION

Adjunct Professor at Columbia University,
Writer and Editor at Nightingale, a Medium.com Publication

Adjunct Professor at Columbia University,
Writer and Editor at Nightingale, a Medium.com Publication