Built In determines the winners of Best Places to Work based on an algorithm using company data about compensation and benefits. To reflect the benefits candidates are searching for more frequently on Built In, the program also weighs criteria like remote and flexible work opportunities, programs for DEI and other people-first cultural offerings.  

“I’d like to extend our heartfelt congratulations to the 2024 Best Places to Work winners,” says Built In Founder and CEO Maria Christopoulos Katris. “I am truly inspired by these companies that have risen to the challenge of fostering a positive work environment, maintaining a strong brand, and ensuring employee satisfaction. The future is filled with promise and we are so excited to see what lies ahead.”

Our 2024 recognitions include:

• Seattle, WA Best Midsize Companies to Work For
• Seattle, WA Best Places to Work
• Colorado Best Midsize Companies to Work For
• Colorado Best Places to Work

At Adswerve, we’re committed to fostering an exceptional employee culture. Our team’s dedication and collaboration are the cornerstones of our success, making these awards particularly meaningful. 
To explore the full list of winners and learn more about the Built In 2024 Best Places to Work Awards, visit the 2024 Built In Awards Winners List.

The post Adswerve Recognized With Four Built In 2024 Best Places to Work Awards appeared first on Adswerve.

1. Prepare For AI
   Artificial Intelligence (AI) is more than a buzzword; it’s a strategic imperative. Agencies should use 2024 to dive into AI, understanding and experimenting with its capabilities to stay ahead of the curve. Establish an internal AI council to oversee the development of responsible AI guidelines, employee training, and data transparency practices. This preparation will help you not only adopt AI effectively but also act as a strategic advisor to your clients, ensuring responsible and efficient AI integration.
2. Embrace Data Privacy
   With significant shifts in data privacy norms, including the phasing out of third-party cookies by Google and new link-tracking protections by Apple, the way we approach consumer data is changing. Agencies must guide clients towards robust first-party data collection and activation strategies, emphasizing the importance of user consent and transparency. Utilizing consent management platforms (CMPs) will simplify adherence to these new norms and regulations, ensuring compliance and fostering trust between brands and consumers.
3. Experiment With Connected TV (CTV)
   As linear TV’s dominance wanes, Connected TV presents a burgeoning opportunity for marketers. Agencies should focus on mastering CTV’s unique challenges, advising clients on creating standout creative campaigns and leveraging new tools for holistic campaign management. Embracing CTV now will enable agencies to capitalize on its growing reach and effectiveness before the market becomes saturated.

For a deeper look at strategic priorities for 2024, dive into the full article on Forbes and if you have additional questions, please contact us to see how we can help.

The post Forbes Agency Council: 3 Strategic Priorities For Marketing Agencies In 2024 appeared first on Adswerve.

Now, let’s explore some of the top areas that can set you up for success in 2024!

Buzzworthy Blogs to Add to Your Strategy Toolkit

Display Media: The Synergy of CM360 and DV360
Discover how the combination of CM360 and DV360 enhances display campaign management, attribution and durable audience solutions.

Search: Paid Search Optimization Tips
Learn four innovative strategies to boost success and stretch your paid search budget further.

Connected TV: Elevating CTV with DV360
CTV is more important than ever. See how DV360 takes Connected TV and YouTube advertising to the next level.

GA4: GA4 Conversion Events Insights
Get a better understanding of GA4’s advanced conversion tracking and how it can be used set up lasting data-driven success.

Adobe Customer Journey Analytics: Migrating to Adobe CJA
Explore key observations around the journey and expectations of shifting to Adobe Customer Journey Analytics.

Trending Topics: Webinars & Guides for Durable Success

Media Trends Webinar: Navigating 2024 with Expert Insights
Gain expert insights on 2024 media trends and digital campaign planning in CM360, DV360, SA360 and Google Ads!

Search Ads 360 Guide: Your Guide to the New Search Ads 360 Experience
Take your search efforts to new heights with this comprehensive guide to navigating the enhanced Search Ads 360 experience.

Search + Social Webinar: A New Way To Unify Search + Social
Learn how to unify Search and Social by harnessing the power of the new SA360 and Adswerve’s UNIFIRE technology.

Google Analytics 4 Guide: Master the Basics of GA4
Master GA4’s top new features and gain valuable insights into best practices for implementation, integrations and more.

GA4 BigQuery Export: Guide to Attribution
Discover how to continue using classic attribution models with the GA4 BigQuery export feature.

What’s Next?

Stay tuned for our 2024 content line-up! Our thought leaders are already crafting new insights in media, analytics, data, and cloud. Visit our Adswerve Resources & Insights page for the latest updates and breakthroughs. Have questions? We’re here to help! Just contact us to schedule a chat!

The post 2024 Readiness: 10 Must-Reads in Media, Analytics and Data Science appeared first on Adswerve.

Adobe has documentation to help with the exercise; however, it can feel a little like being thrown into the deep end of the pool. Additionally, web and mobile digital analytics data can be a unique beast that can cause even more complexities. With this article, we hope to equip you with the knowledge and best practices to help you be a good steward and maximize the value of your organization’s digital data in Adobe Experience Platform.

What is Data Modeling?

A data modeler designs databases to efficiently access and use data for business processes. It may not always seem clear in AEP’s self-service user interface, but under the hood, this is the role enabled by XDM. In fact, with support for objects and nested arrays, the resulting tables can be complex enough to give even the most experienced data modeler heartburn. Without standards and best practices, AEP’s data lake can quickly become the dreaded data swamp.

XDM Schema 101

Adobe has taken a unique approach that doesn’t fully adopt some of the traditional principles of modeling data with SQL. Instead, they have applied a framework called XDM. To understand the XDM schema, we’ll first need to cover some key terms. 

At the most basic level, a schema is a defined list of fields. The fields should combine to describe a real-world object or experience as a data structure. Additionally, it serves as a contract to guarantee that incoming data follows a consistent and defined format with basic validation for each field (data types, lengths, etc.). 

AEP additionally applies the concept of a class to assign schemas a set of required fields that inform Adobe products whether the data contained within that dataset will be compatible with that product. For example, Customer Journey Analytics (CJA) expects at least one dataset with a schema class of XDM Experience Event due to the required timestamp utilized for reconstructing journeys. Classes are the primary way to identify the three main types of datasets expected to exist in AEP:

• Event
• Profile
• Lookup

Building a Schema with Field Groups

Also, unique to AEP, fields are created inside field groups to encourage the reuse of commonly combined fields and to standardize schema creation. You may see documentation or API references to “mixins”, which was an early name for field groups that was eventually replaced in the user interface. 

Adobe has made a number of pre-built, standard field groups available. While these field groups may help with some very basic attributes that you will collect along a customer journey, they are not comprehensive and will miss critical fields needed for many web and mobile experiences (e.g., forms, navigation, errors, etc.). Additionally, they will often contain extraneous fields that will be unnecessary for your implementation. As a result, almost every implementation will involve a considerable amount of custom field group creation, which is where the bulk of your newfound data modeling skills will become handy.

Standard or Custom Field Groups

The hardest decision point when designing XDM schema is if and when to leverage the Adobe standard field groups. What are the differences between standard and custom field groups?

Standard Field Groups 

• Preloaded by Adobe and made available as part of their vision to create a unified data model
• May enable automated functionality (e.g., device lookups on Datastreams via Edge Network)
• Appear at the root level of a schema

Custom Field Groups 

• Manually added by an Adobe customer
• Appear within a “_tenant” namespace object at the root (where “tenant” is replaced with your organization’s name)
• Allow full control over which fields to include, as well as ensuring that naming matches internal terminology and semantics

Additionally, there is an ability to extend a standard Field Group with custom fields, creating a hybrid field group. This approach allows you to add custom fields to standard objects without having to use the “_tenant” namespace. 

It can be helpful to think of field groups as data building blocks that can be stacked to create schemas. In the below example, we have a retail experience in which a customer can navigate a website and make purchases. There are two datasets intended to be used in Adobe CJA: website clickstream data and transactional data from a back-end order management system.

You can see a standard Commerce Details field group block that exists across both schemas. The common field group is essential for reporting on fields across these datasets in a continuous way. For example, a use case may be to analyze whether products with certain behaviors on the website (e.g., viewing a video, reading reviews, etc.) are less likely to cancel or return an order. If the schemas are constructed with mismatched blocks, they won’t be able to correlate the product activity from the website to the outcomes in the transaction dataset.¹  

 1 There are options within Adobe CJA to work around the ‘mismatched blocks’ scenario with Derived Fields, but it should only be considered as temporary until the data model can be updated appropriately.

Planning Ahead to Avoid Breaking Changes

As you dive into creating your first schema, it is important to be aware of breaking changes and their impact on schema design choices. As your business grows, your schema can and should grow, too. Any modification that could cause existing data stored against the schema to become non-conforming is considered a breaking change and is strictly not allowed. If you haven’t collected data yet, you can make any changes you need. Once that dataset is populated, you will be subject to breaking changes, which include, but are not limited to:

• Field path changes, including field name (Display name can change at any time)
• Data type changes
• Field deletions
• Changing an optional field to required

If you need to make one of these breaking changes, you will have to create an entirely new schema and reprocess your data to conform to the new schema—a painful process no one wants to endure. For that reason, you will want to be deliberate about additions to your schema, as once data is stored against it, there are limited options for going back.

Data Modeling Best Practices

While following standards and best practices provides a foundation for governance and compliance, it’s just as much about improving the user experience for data consumers. The more uniform and consistent the data is, the easier it is to explore and contextualize. Just as you want a digital experience to be effortless and intuitive to use, working with data should be no different.

Now that you have a basic background on AEP schemas let’s get down to some best practices. As this guide is more focused on web and mobile behavioral data, these recommendations will be aimed at those types of data structures. That said, a lot of these principles do apply to any type of data in AEP.

• Use Adobe-owned field groups as a starting point. The standard AEP Web SDK ExperienceEvent (web and mobile apps) and AEP Mobile Lifecycle Details (for mobile apps) field groups align to automated functions within Adobe’s data collection tools. If using AEP Data Collection, take advantage of those and use them as a starting point for your web and mobile datasets. The specific objects that will be populated by the automated collection, when enabled, are:
  • placeContext (Web SDK)
  • environment (Web SDK and Mobile Lifecycle)
  • device (Web SDK and Mobile Lifecycle)
  • application (Mobile Lifecycle)
• Within the standard field groups, don’t populate them just because they’re there and remove them if you won’t use them. If you find that one of the standard field groups aligns with your data needs, there will often be excess fields that you don’t need for your use cases. There is an implementation and maintenance cost for every field you choose to populate. Pick the fields that provide the most value for your business and remove the rest. 
• Don’t sacrifice semantics to adhere to the standard field groups. If it isn’t clear to you what a standard field is supposed to be and it isn’t auto-generated, don’t use it. Opt for custom field groups or extend the standard field group when a standard field doesn’t make sense for how your business talks about the data (e.g., create a custom cartAdditions instead of using productListAdds). As long as a field isn’t system-generated, it will benefit your users to have clear, understandable fields instead of ambiguous and force-fit fields to meet the standard. 
• Don’t overload a single field with multiple types of data in delimited or concatenated strings. Normalize the data in separate fields or in an array if it’s a list. This will ensure that policies can be applied at the field level and data is protected appropriately.
• Don’t create website, app or dataset-specific field groups for schemas. Organize field groups by the real-world objects and entities they are meant to represent. Those will likely repeat across your schemas and reusing common field groups will ensure that data aligns across datasets and that schema updates cascade to each of your schemas.
• Create one large schema for web and mobile datasets, similar to a Global Report Suite. There can be exceptions to this, but generally, having fewer schemas to maintain will be better. Additionally, keep your license structure in mind. For example, CJA Foundation currently allows only one event dataset per Connection. Creating a separate schema and dataset for every site may prevent you from bringing all of the data together in CJA.
• Don’t create field groups for a vendor/tool. Instead, organize them by the generic real-world object or experience they are meant to describe (e.g., leads, marketing campaigns, appointments, etc.). This will allow you to reuse them across schemas and datasets.
• Make sure that commonly occurring fields (e.g., SKU id, location id, campaign id) share the same field group in their respective schemas. If they are in different field groups and have different schema paths, the fields will mismatch when you attempt to merge them as one continuous journey.
• Do NOT create flex fields. Along the way, you may hear something to the effect of “We don’t always know what we need to collect, so can we create an array that can hold any field and value combination?” DON’T. This may seem like a clever way around having to create new fields, but it is a notorious source for data leaks. One of the purposes of schema is to label and protect fields with policies appropriate for the sensitivity of the data within it. Once a field has potentially endless meanings, it will inevitably be misused. Additionally, it is fragile and time-intensive for data consumers to have to apply logic when reading from these types of flex fields.
• Carefully consider whether to append data to a schema or use a separate lookup schema. When building out your data model, sometimes you will have data that can be captured and stamped directly on an Experience Event or be loaded in through a backend process and enriched with a lookup dataset. Be intentional about whether you choose one or both of these options. Some factors to consider:  
  • Fields that can change from day to day or user-to-user (e.g., price, wait time, loyalty status) are a better fit to be stamped to the Experience Event.
  • Slow-changing data like product information or user demographics will be a better fit for utilizing lookups to enrich the event data. The downside of a lookup is that if data does change (taxonomy, for example), the historical representation of what it was when the event happened will be lost. If that matters for your use case, stamping it to the event may be more appropriate.

Naming Conventions 

When it comes to naming conventions in data modeling, the most important thing is that you have one, regardless of the details of what it is. The conventions below primarily take a nod from the out-of-the-box field groups provided by Adobe while applying some other common conventions to help with readability. Here are some suggestions, but feel free to add on with your own:

• For field Name, use camel case (cartAdditions, mediaPlayer)
• For field Display Name, use normal spacing and capitalization (Cart Additions, Media Player)
• For field Name, don’t repeat the name of the parent object in a child field to keep schema paths readable.
• For field Display Name, include the parent object name to make it easier to find in field lists.

• Don’t use acronyms except for the most commonly used ones (sku, id, ip, etc.)
• Don’t use internal project names.
• Don’t use vendor or tool names except for vendor-specific IDs.
• Reserve plural field names (e.g., filters, payments, checkouts) for two types of fields:
  • Arrays: whether an array of objects, strings, or other data types, plural names help data consumers understand that they are working with multiples of that field.
  • Metrics or events: a convention sometimes used in Adobe field groups, these will typically be an object with two fields inside: ‘id’ and ‘value’. The ‘value’ would contain a numeric value to increment the metric, most of the time 1, while the ‘id’ would serve as an optional identifier for deduplication. 

Final Thoughts

Data modeling with XDM can be overwhelming. To make it more manageable, break it into pieces. Do a trial run of your data models with sample data, and it can help you see how they come together. Most importantly, know that there is no perfect data model, but with a good foundation and focus on semantics, you can ensure that your organization’s Adobe CJA use cases can be met.

This article is the first in a series to help you lay the foundation you need to be successful with Adobe Customer Journey Analytics. Additionally, we geek out about data models and would be excited to schedule a free 30-minute consultation to discuss your implementation of Adobe CJA.

The post Adobe CJA Foundations: Data Modeling with XDM Schema appeared first on Adswerve.

As we approach the start of a brand new year, we are excited to guide you in maximizing your usage of the new Search Ads 360. To elevate your digital advertising efforts in the upcoming year, we have compiled a list of five enhanced or new features that deserve your attention.

5 Exciting Features That Will Impact Your 2024 Strategy

1)Value-Based Bidding 

Although forms of value-based bidding (VBB) have been around for quite some time, in the new SA360 experience, bidding to value on counter-based conversion actions has never been easier to configure. With VBB in the New Experience, you can focus on capturing the actions and customers that matter most instead of just acquiring leads at a set cost.

The Power of Floodlights

If you’re already harnessing the power of Sales Floodlight activities, additional configuration isn’t necessary but may be relevant if you’re missing out on offline conversion data. If you aren’t using Sales Floodlights, all you need to do in the New UI is assign a performance-based monetary “value” to one or more conversion actions. 

Tips for Determining Conversion Value

While assigning the value is easy, determining the value takes some thought. Here are some methods to consider:

1. Lead Value: Google’s Conversion Value Calculator can be utilized to calculate the value of each lead.
   
2. Lead-based Profit Margin: If your CRM reports that 30% of leads become customers, and the average order value is $350 with a profit margin of 60%, the conversion value would amount to $63 (350 x 0.60 x 0.30).
3. Estimated Lead Lifetime Value: If it is determined that over their lifetime, a customer spends on average an additional $1,000, and the profit margin remains the same, the profit per customer would be $810 (($350 + $1,000) x 0.60). With a conversion rate of 30%, the conversion value would be $243.

Enhancing VBB through Custom Variable Adjustments and Conversion Value Rules

In addition to defining the value of a conversion action, you can enhance VBB even more by employing Custom Variable Adjustments or Conversion Value Rules in your campaign settings. 

• Through Custom Variable Adjustments, you can optimize VBB by bidding towards specific custom variables passed through your floodlight tags, such as travel and location, loyalty, products, and more. This allows you to consider additional factors that are important to your business, in addition to the base value of a conversion. 
• With Conversion Value Rules, you can fine-tune assigned values based on the user’s geographical location, device, or audience. This can be helpful if you have a national campaign but want to be able to bid more aggressively towards higher converting areas. In this scenario, you can assign different values to customers based on their geographical location. 
  • For example, customers in New York may be worth $500 to your business, while customers in Los Angeles are valued at $300 and customers in smaller metros like Kansas City are valued at $100. By creating value rules based on location, you can tailor your bidding strategy to prioritize specific areas and optimize your return on investment. 

2) GA4 Linking

While GA4 property linking was initially available in the Legacy SA360 UI, capabilities have improved for advertisers who have migrated to the new experience. In the previous UI, only one GA4 property could be linked to each SA360 advertiser. In the new UI, each SA360 sub-manager can be associated with up to 200 GA4 properties. Additionally, the New SA360 UI now supports auction-time bidding with GA4 conversions for Google Ads campaigns and intra-day bidding for Microsoft Ads campaigns. In 2024, GA4 data-driven attribution will also be rolling out to the New SA360 UI. 

3) Performance Center 

Another feature that has been completely revamped is SA360’s Performance Center, which now combines the best parts of Performance Planner from Google Ads with the legacy Budget Management feature to provide an improved planning, forecasting, and budget optimization suite. 

With this tool, your team can plan, optimize, and report on ad spend for the fiscal year, quarter, month, or week. When combined with one of SA360’s robust budget bid strategies, you can supercharge your budget and efforts by automating your performance plan to not only hit your desired spend target but bid towards a set performance goal. 

4) Additional Campaign Support

Back in 2018, Adwords was officially rebranded to Google Ads in order to better represent all available creative formats. Google hasn’t looked back since then, continuously updating its landmark Search platform by introducing new campaign types and features. More recently, Google transitioned Smart Shopping campaigns into Performance Max, uniting search, shopping, display, and video formats in one powerful configuration. 

With all of these changes, SA360 needed to adapt. The New SA360 UI now supports the creation and management of the most commonly used campaign types in Google Ads, including Performance Max, Demand Gen (Discovery), Display, and Video (YouTube). 

The new SA360 now offers a variety of additional campaign support, including creative management through the integration of Google Ads’ Asset Library. And Performance Max campaigns can now employ SA360 bid strategies optimizing towards Floodlight activities, which is expected to roll out to other campaign types in the future. You can even include a Performance Max campaign within a Performance Plan, and manage its spend with a Budget Bid Strategy, allocating spend between Search and Performance Max campaigns based on current and historical performance. 

5) Looker + New Reporting

If you have experience with Google Ads reporting, you’ll easily adapt to SA360’s new reporting interface. The New SA360 offers 77 pre-defined reports that can be customized and utilized at all levels of the New SA360 hierarchy. If none of the available templates fit your reporting needs, you can also build a custom report from scratch. Although Executive Reports have been deprecated, executive-level reporting hasn’t gone away. At the Manager account level (previously Agency), any Custom Columns with manager-level ownership can now be added to reports, giving users the ability to see overall account performance with greater precision. 

Speaking of reports in the Legacy UI, if you were using Conversions & Visits reports to filter out Scaled Conversions modeling, that report has now been split into two within the New UI. The Conversion events report outlines every attributed conversion event that has occurred, detailed with a timestamp, click ID, and attribution information. Like the Legacy Conversions & Visits report, this report excludes Scaled Conversions. The Visit events report is very similar, but focuses on visits instead of conversions, and provides greater granularity. Both of these reports must be downloaded, as several dimensions aren’t supported within the UI.

The New SA360 UI also incorporates dashboards, allowing you to easily combine multiple reports all within a user-friendly navigation. If creating an impressive visualization is your goal, however, Looker Studio might be a better option. By tapping into SA360’s new Looker Studio data connector, you can present performance data in highly customizable charts, scorecards, and tables. Unlike the Legacy SA360 data connector in Looker Studio, the new connector allows you to pull in custom conversion and formula columns for increased precision in conversion reporting and attribution. You can even blend SA360 data with other channel data in Looker Studio to get a better picture of how your various marketing channels are working together to drive performance.

Honorable Mentions 

Facebook Gateway

Another exciting update to the new Search Ads 360 is the incorporation of Gateway technology. The current state of this feature allows Facebook Ad accounts to ultimately sync into SA360. However, even with the Gateway feature available, there is a significant technical and resource lift in order to enable the feature properly and get vital social data in Search Ads 360. That is where Adswerve’s patented technology, called UNIFIRE, comes in.

Our UNIFIRE app simplifies the complex process through UNIFIRE’s simple UI; only a few button clicks are needed to automate the Gateway creation process, a process that typically calls for IT or a developer. Instead, UNIFIRE enables you to fast forward to creating your Facebook Ads client account in SA360 and begin reaping the benefits of SA360’s tried and true auto-tagging, reporting, and management capabilities.

Recommendations Tab

In recent years, Google and Microsoft have been putting more emphasis on Optimization Scores, which Google describes as “an estimate of how well your Google Ads account is set to perform.” While the exact impact of the Optimization Score on performance is unknown, Google notes that a score of 100% means that an account will be able to perform at full potential. Optimization scores can be improved by either manually or auto-applying system recommendations, which are now available for both Google and Microsoft Ads in the New SA360 UI. When scoped down to the Google Ads client-account level, you’ll be able to see your current Optimization Score and the percentage improvement that accepting specific recommendations will yield. At the sub-manager level, recommendations can also factor in Floodlight conversion data if Floodlights are shared with Google Ads and applied as a conversion goal.

Interested in learning more about the powerful potential of the new Search Ads 360? We’d love to chat! Contact us today.

The post The New SA360: 5 Features That Will Impact Your 2024 Strategy appeared first on Adswerve.

Solutions

We don’t know how many segments we need or how many will be useful.

We can think of this as a model optimization problem where we’d want to tune hyper-parameters. In such a case, we can run a k-means model within a hyper-parameter optimization framework (Optuna, hyperopt, vizier).

Hyper-parameters for k-means could be:

• Number of clusters
• Number of iterations
• Tolerance

The evaluation metrics could be:

• Silhouette score
• Mean Squared Distance
• Davies–Bouldin Index (DBI)

However, optimization for only one of those metrics might have some side effects. Usually, the more clusters there are the better the metric score. As clusters get smaller, usually they are more cohesive and better separated, so our optimization would likely return a very high number of clusters as our best choice.

Most of the time, we prefer a small number of clusters because they are easier to explain, and there is typically less business value in having many small niche clusters. Enter multi-objective optimization.

An interesting approach is to have two optimizing criteria:

• Maximize the evaluation metric, such as the Silhouette score
• Minimize the number of clusters

That way, we are trying to get better clusters (by evaluation metric), but at the same time, we have as few of them as are still viable.

Multi-objective optimization does not return one best solution but a Pareto front of best solutions. From there, we still have to pick one to go with. However, now we have an excellent view of what the best evaluation metric is for each number of clusters. We could have a heuristic saying, “Take the model with the number of clusters equal to X, but that model shouldn’t be worse than Y% from the best-scored model based on the evaluation metric.” 

It’s like saying we don’t need the best model because it also comes with many clusters that may not be relevant, so we are willing to trade a little bit of performance for a smaller number of clusters.

The multi-objective optimization can be freely expanded into even more optimization targets. You should have as evenly sized clusters as possible. If so, use the Gini coefficient to measure the size inequality and minimize it.

It could get even more interesting if the optimization process includes feature selection, meaning you want to select only features that generate the best possible results. When optimizing for feature selection, you don’t want to drop many of them, so you might also consider optimizing to maximize the share of features used.

Result of the multi-objective optimization. Three optimization targets produce a 3D Pareto-front.

Once you have the segments, they need meaningful business names.

There is no automated naming solution (yet), so you will still need human intervention to come up with names.

One interesting alternative worth exploring would be using Large Language Models (LLMs) to recommend name ideas based on the cluster centroids given. Eventually, this could also be automated.

Segments need to be explainable.

The easiest way to make segments explainable is to reduce the number of attributes and make them of the same ‘domain.’ Even as low as — or below — 10 attributes can make a world of difference, making it easy to understand what is happening.

Reducing the number of attributes is usually relatively easy if you pick a direction (or  ‘domain’) you want to cluster on. Some examples:

• Interest
  • Possible features: product categories visited
• Engagement
  • Possible features: time on site, number of page views, number of visits, frequency of visits
• Sales funnel stage
  • Possible features: Key steps of the funnel visited (add to cart, checkout, product page visit, purchase)

You can always cluster on those separately and still slice and dice different segments from different models. In other words, by not throwing everything into one model but separating it into different models, we retain all possibilities, likely with better explainability and flexibility.

Consider our example from the beginning. We want to come up with segments based on interest. How do we develop attributes for such a model with GA4 data? Typically, websites have a structured way of setting up URLs (page paths). This means that, in most cases, we can extract at least a product category from the page path and use the number of times someone has visited that product category as an attribute.

An example:

Visitor 1

/womens/running/shoes ~ 1x
/womens/tops-and-t-shirts ~ 4x 
/accessories/basketballs ~ 8x

Features:
running_shoes: 1; tops_and_t_shirts: 4; basketballs: 8

Visitor 2

/womens/tops-and-t-shirts ~ 1x 
/accessories/basketballs ~ 2x

Features:
running_shoes: 0; tops_and_t_shirts: 1; basketballs: 2

Engagement-wise, those two visitors are very different. One has 13 page views, and the other only three. Interest-wise, though, they seem very similar and should likely be in the same segment. 

We also don’t know which one of those product categories is more visited in general, which may mean that going there often isn’t an expression of interest because most visitors go there, too. Or, some product categories might be visited very sparsely, so going to those is a much bigger expression of interest.

To solve the product category importance issue, we can use a preprocessing technique called TF-IDF, often used for word importance in text retrieval. We just need to pretend each product category is a word, and each visitor is a document (as in containing words). The process will weight each category accordingly based on the frequency of its visitations.

And to ensure both visitors end up in the same cluster, we should neutralize their engagement levels. We can accomplish this by using cosine distance as the similarity metric instead of the classic Euclidean distance. We could also address it by normalizing each visitor’s vector to have a distance of 1 (L2 normalization) and still using the Euclidean distance. Both ways assume a k-means clustering model, though it would also work with other similarity-based models.

Retraining the model might generate different segments.

This is addressed by ensuring random seeds in the code are appropriately set, meaning each run will produce the same outcomes. This might not be perfect if data changes quickly or a new model is retrained on data that doesn’t overlap well with the model’s initial training. Either way, setting a random seed is always recommended.

Another solution would be to use the prior model’s centroids as starting points for the new model. Essentially, this mimics a continuous learning process where the start of a new model’s training is exactly where the old model’s centroids were. Together with the random seed, it ensures that the segments won’t change dramatically with each retrain.

This solution also opens up an opportunity for efficient model monitoring. When each model is trained, you can easily measure the difference between centroids from one model to the next. Once large enough, this model drift could become a trigger to redefine the clusters or train a completely new model.

Conclusion

We can imagine having a well-done clustering model that would include training and prediction pipelines and could run continuously by implementing the ideas discussed here. We can also implement model monitoring to detect better when fresh retraining is needed instead of continuous training. 

Ultimately, you can deliver solid segments to the GMP for activation purposes, so the great data science work doesn’t just stay within the warehouse but can be used and acted upon in your campaigns.

Questions? Want to learn more? Contact us to start the conversation.

The post Using BigQuery and GA4: Thoughts on Audience Segmentation (Part 2) Solutions appeared first on Adswerve.

“When employees feel they belong, they are more engaged, motivated and better equipped to bring their best selves to work every day.”

Courtney Reed, DEI Manager
From Built-In: How Connectivity, Community, and Creativity Power Adswerve’s Virtual-First Culture

Sharing our story and culture – like our “three Cs” of connectivity, community, and creativity – was so fun! Our “three Cs” principles are not just buzzwords; they ensure that our team members feel linked, regardless of location.

“Working from home has provided flexibility around my schedule and facilitated greater autonomy over my time.”

Nílber Remón, Data Engineer
From Built-In: How Connectivity, Community, and Creativity Power Adswerve’s Virtual-First Culture

We encourage you to check out BuiltIn’s recent article for additional perspective from Adswerve CEO Roger Berdusco, plus insights from data engineer Nílber Remón and DEI manager Courtney Reed on remote work-life and the benefits in terms of flexibility and autonomy, particularly in technical roles. 

Interested in joining our team? Check out our Careers Page and don’t forget to follow us and engage on LinkedIn.

The post Leading the Virtual Way: Built-In Spotlights Adswerve’s Connected Culture appeared first on Adswerve.

Google’s Display & Video 360 (DV360) comes with a transparent cost, so media buyers want to understand the value they receive in exchange. The answer to the question is with another question, “What do your media buying and team goals look like?” Read below to see five compelling cases in which DV360 can be a better fit for your business.

5 Compelling Cases to Make the Move to DV60

1. Campaign Manager 360 (CM360) is Already in Your Tool Repertoire 

DV360 natively integrates with CM360 for creative placements and floodlights. This eliminates the need to export CM360 tags and reupload them into another platform, piggyback additional pixels for measurement or first-party audience collection, and brings your whole DV360 strategy into CM360 alongside the rest of your media for streamlined attribution reporting. Who doesn’t love maximizing tools? 

2. Media Workflow and Cost Efficiencies Are Top of Mind

An additional benefit to the note above is how your teams can work more efficiently and save operating costs since CM360 and DV360 are linked. You and your team have more time for media planning, reporting and optimizations, all while saving room for potential errors or asking additional teams to place more pixels. Even if you don’t have CM360, DV360 simplifies workflows as you are able to buy multiple ad formats in one place, so your teams don’t need to communicate, traffic, build, report and project manage multiple vendors to effectively execute a campaign. This time savings leads to cost savings, which reduces the all-in cost of using DV360.

3. Your Campaigns Involve Many Formats and Inventory Types

Speaking of buying multiple ad formats in one place, DV360 seamlessly allows media buyers to access Online Video, Connected TV, YouTube (YouTube Shorts and YouTube TV included), Display, Audio, Digital Out of Home (DOOH), and Native in one UI. Plus, buyers are able to go beyond Google’s O&O inventory, so your team can buy the whole gamut of media in one place. This includes 57+ media exchanges like Magnite, Rubicon, AdsWizz, FreeWheel, GumGum, Kargo, Place Exchange, ShareThrough, TapJoy, and Vistar. Not only does this consolidation lead to workflow time savings, but DV360 allows buyers to layer on frequency caps across all formats so your media dollars have less waste and go further, too.

4. Detailed reporting and control are a priority

We’ve heard from many clients that they are okay with the limited display inventory in Google Ads, but they would like to dive deeper into the platform results to confirm its backend efficacy. DV360 is likely to better fit your needs in this scenario! You can get granular insights into common metrics like views, impressions, clicks, conversions (click through and view through), viewability, and more complex metrics like quartile and completed views, % on full-screen, CPCV, media cost, and audio mutes.

There are also numerous dimensions to see these metrics by, like line item, audience segment, ad group, app/URL, creative, creative type, device type, exchange, and time of day. The best part is you can turn these insights into actionable edits to your campaigns, as most of these dimensions correlate to different setting opportunities in DV360. 

For clients who like the “set it and forget it” method, where there is little control or decisions needing to be made on your part, Google Ads can be successful for you. For those clients who know digital media is nuanced and want to have control over where their dollars are spent and then be able to report out accordingly to track it, DV360 is in your wheelhouse! The DSP lets buyers set budgets, bid strategies, and targeting decisions at the granular level of a Line Item. This allows campaigns to be more dynamic and effective at reaching who you need but also optimizing specifically for them. 

5. You and your team like maximum targeting opportunities

If you know diversity is critical to a media campaign’s success, even if running a simple prospecting or remarketing campaign, then DV360 will fit you like a glove. Consumers are all over desktops, phones, Connected TV devices, and just generally out and about in the world constantly connected. This creates a need to reach them everywhere. To do that, you need a robust DSP – not just a Google silo-ed Google Ads. DV360 is equipped with the following targeting selections to let you slice and dice how you need:

• Third-party data (custom-built, B2B segments, shopping behaviors, etc.)
• GA4 audiences
• CRM onboarding with LiveRamp (and other DMPs) 
• Brand Safety + Viewability (including IAS and DV wrapped tags and pre-bid segments)
• 57+ exchanges, including those that specialize in OLV, Native, CTV, and DOOH
• Deal IDs to serve on specific-publisher inventory
• Exclusion audiences (layer on each Line Item and set specific timeframes for each)

If you have any additional questions or would like to chat more about how DV360 could be a great fit for your business, please contact us today!

The post Is DV360 Right For You? 5 Compelling Cases to Move Beyond Google Ads’ Display Network appeared first on Adswerve.

This significant milestone follows our one-year anniversary of acquiring Metric Partners and advancing to silver status in July. Our swift progression from bronze status at the beginning of the year underscores Adswerve’s dedication to enhancing our Adobe services and our unwavering commitment to driving exceptional results for our clients.

By earning Analytics Specialization, Adswerve has also proven its expertise and success in providing consistent customer success in Analytics with Adobe. 

“Adswerve is the only premier digital consultancy at the intersection of media, data and technology, and we are committed to helping marketing and tech leaders enhance the value of their martech investments and forge deeper connections with their customers,” stated Charles Farina, Adsweve VP of Innovation. Our forward-thinking approach to data and analytics strategies leverages the strengths of both Adobe and Google platforms, ensuring that our clients achieve and surpass their marketing goals.”

Adobe Gold level cements our status as a premier enterprise partner in data and analytics. We look forward to expanding our catalog of specializations as more customers choose Adswerve to strengthen their Adobe solutions and strategies. 

“Our forward-thinking approach to data and analytics strategies leverages the strengths of both Adobe and Google platforms, ensuring that our clients achieve and surpass their marketing goals.”

Charles Farina, VP of Innovation

If you are considering Adobe or looking for a trusted partner to ensure you’re getting the most out of your tech investments, we’re ready to help! Reach out and one of our experts will be in touch shortly.

The post Adswerve Achieves Adobe Gold Partner Status with Analytics Specialization appeared first on Adswerve.

Watch now to discover key insights, including:

• Media Channel Priorities for 2024
• Strategies for CM360, SA360, DV360 and Google Ads in 2024
• Client and Industry trends
• Data Privacy Tips & more!

Additional Resources:

As mentioned during the webinar, more details on tracking U.S. state consumer data privacy legislation can be found at Husch Blackwell’s 2023 State Privacy Law Tracker.

Media Trends Hotline: Navigating 2024 with Expert Insights
Watch Now On-Demand!

The post On-Demand Webinar: Media Trends Hotline: Navigating 2024 with Expert Insights appeared first on Adswerve.