Damon Torgerson

Jeff Bezos interview in Wired

Wed, 28 Nov 2012 23:11:19 -0800

Jeff Bezos interview in Wired:

If you haven’t read this interview in Wired last November…

Live in the future, then build what's missing

Wed, 21 Nov 2012 18:32:51 -0800

Live in the future, then build what's missing:

Another great essay on startups by Paul Graham

Use Social Contracts to Help Focus Your Bootstrapped Startup

Wed, 21 Nov 2012 10:25:50 -0800

A little over a month ago, I joined a local startup community called Founders in Motion started by Hakon Verespej. Seattle has a vibrant startup community and any given day you can attend a startup coffee in the morning, grab lunch with startups, and learn about starting up that night. There are many, many opportunities to meet and learn about startups.

These events are great as being in a startup can sometimes feel a little lonely. One of the things I appreciate about Founders in Motion is that we foster accountability with each other. Each week, I make a commitment to the group about my next steps. I get valuable feedback on the steps but what I value most is that I make a commitment to drive my business forward. I find it very easy to justify my strengths when making decisions about next steps to take with my startup. As a technical cofounder, there’s always something technical that needs doing. I’m a big believer in the lean startup approach, which tends to favour talking to customers rather than building on assumptions. I made that mistake in spades on my first startup and so I am always looking for ways to avoid repeating this mistake. It was making a commitment to the group that helped me push past my comfort zone and ask two early adopters to provide warm introductions and to my pleasant surprise, they were more than happy to help.

There are many examples that demonstrate that a social contract can be much stronger than a legal contract. I imagine that if we had venture funding I could use that to help me focus on the business at times when I find myself building on assumption but as we’re bootstrapping, making a social contract helps me focus on executing the business. If you’re a technical founder and you find yourself building on assumption when you should be engaging customers, I’d encourage you to join Founders in Motion and see if making a social contract can’t help you.

California Unified Districts according to 2010 Census Data...

Wed, 14 Nov 2012 08:42:47 -0800

California Unified Districts according to 2010 Census Data thanks to @mapbox and @openstreetmap #opendata

Starting into my second month with GTD

Wed, 07 Nov 2012 10:29:21 -0800

I’ve been more or less adhering to Getting Things Done using the Secret Weapon and Evernote for the past month. So far, it’s been a success and I’m finding myself more organized without having to be more organized.

One thing I struggled with is unifying my various email accounts. For a while, I had Action Pending, Waiting, and Reference folders for each account. As per the TSW’s guidance, I initially consolidated and sent everything to Evernote. However, I was uncomfortable with that because I felt that I couldn’t identify the emails I had to take action against so I went back to three folders…across all of my email accounts. For me, that was nine folders I had to visit plus Evernote, which was manageable but clumsy. After consideration, I decided to follow TSW’s guidance and consolidate everything into Evernote. It’s not ideal, but it gives me one place to work from rather than multiple lists.

I still haven’t gone through everything in my life and ordered it according to GTD but the changes I have made have provided me with more time. It’s an evolution.

If you practice GTD, do you have any tips to make the transition better?

Small Starts

Tue, 06 Nov 2012 19:44:32 -0800

A few days ago, our student transfer application surpassed 5000 transfer requests. Having worked on many enterprise systems, 5000 is not a large number but one of the things that I’ve always found challenging with enterprise systems is that the benefits are often difficult to quantify and due to the bureaucracy of organizations challenging to realize. This has not been my experience working with school districts.

Before using our application, parents had to fill out a paper form and drop it off at the school district. District personnel then keyed in the request and maintained the information in a combination of Excel documents, paper folders, and local computer files. As you can imagine reporting was time consuming and managing overall the student transfer process required a high degree of institutional knowledge.

Now parents complete the form online and district staff manage the entire process in a secure, open standards based, web application that requires minimal training. Reporting is built in and the entire process is very transparent.

All told, our student transfer application cuts processing time down by about 10 minutes per request. So far, this is over 800 hours saved and this speaks nothing of the benefits easier analysis provides and a (much) more transparent process.

If you build enterprise-y applications for a living these numbers might not impress you. However, it’s among the most rewarding work I’ve ever done. Our application is used on a daily basis, our customers are delighted, and I’m making a (small) difference in something I care about. It could be a lot worse.

I love Evernote and cannot wait for version 5

Tue, 30 Oct 2012 11:25:26 -0700

I love Evernote and cannot wait for version 5

Planet Money's Fake Candidate

Sun, 28 Oct 2012 14:37:00 -0700

Planet Money has a fun series that results in the creation of a fake presidential candidate based on economic policies with (generally) broad support across (e.g., left to right) the economic community…that no politician would touch. Then they get political consultants to help them sell the policies.

The episodes (in order)

The ads

Dipping a toe into value proposition validation

Fri, 26 Oct 2012 18:01:00 -0700

While having dinner with a friend and talking about validating business models, he suggested that I not trivialize the affect of getting customer feedback. If you really care about your idea, it can be a challenge being told it doesn’t solve world hunger. Up until that point, I hadn’t even considered how I might feel about it. I’d be lying if I said it wasn’t in the back of my mind but now as I’m getting my first tests out the door I think it was good advice. I think I’m better prepared mentally to accept the inevitable criticism of my idea and not take it too personally.

My first experiment on validating the value proposition is a survey to school district personal through the CETPA mailing list. Last week at CETPA, I learned that technology directors (another hypothesis) favor email and, in particular, the CETPA listserv. I’ve been monitoring the CETPA listserv for a few weeks and a lot of sharing occurs there. I’m used to LinkedIn, Stack Overflow, and Google Groups so this is a little new (old) for me. I’m very interested to see the results of the survey and the results of the engagement through the list.

My next experiment, which I’m running concurrently next week, will be warm introductions from our existing customers. The survey is a good warmup for the conversations I need to have next week. It will help set my expectations when I’m told I’m not curing cancer.

Why Preschool Can Save the World

Thu, 25 Oct 2012 20:00:00 -0700

Why Preschool Can Save the World :

Preschool provides a 10% annual return on investment.

Attending CETPA and Meeting with Customers

Wed, 24 Oct 2012 22:05:31 -0700

Last week I attended the California Educational Technology Professionals Association annual conference in Monterey. It was a quick trip but well worth it. For our business, it’s very important to understand the technologists view as they are often key partners and sometimes our customers. They don’t always make a decision but they certainly can end it.

In addition to our work being highlighted by one of our customers, I attended two sessions to learn more about the procurement process: Building Meaningful Partnerships Between Business and School Districts; and School Funding in California. The sessions offered a lot of practical advice but what I really liked is that most of the discussion resonated with my values. Districts truly view vendors as partners with the goal of the partnership to improve student outcomes. It’s not that they’re looking to sing kumbaya around a campfire, but if you’re in it for the quick sale, you’re in the wrong industry. To me, favour founders over HAL’s education marketing channel vertical.

I also had an opportunity to meet with one of our customers and gain insight into what drives his decisions (student outcomes). The cynic in me snorts a little and says, “yeah, right, it’s for the children,” but in candid conversation after candid conversation, these folks really are in the industry to leave a positive mark on education. It was a good, quick trip, that provided me greater insight into my customers and helped inspire my first experiment…more on that soon.

Think bigger

Tue, 16 Oct 2012 22:02:00 -0700

Consulting is a reality in my bootstrapped education startup. I have bills to pay and I’m not 20 anymore. I have a lifestyle. It ensures cash-flow and to provides an opportunity to experience the industry. Lately however, I’ve been focused too much on projects and not enough on the startup. I’ve focused too much on the trees and not enough on the forest. Too much on the lifestyle. Worse, I fell into the trap of extrapolating a point into a line and forgot the lean startup principles I always ensure my fellow entrepreneurs remember. Recently, some very close friends encouraged me to think bigger. It wasn’t easy to hear at first but it was a wake up call to get going.

Tomorrow I’m attending a day of the CETPA annual conference where I’ll get a chance to (re)start validating the value proposition and the customer segment of what I think is a really big problem. Of course, if it’s really as big of a problem as I think it is and these are the customers, they’ll tell me. Get out of the building.

The importance of taking a break

Mon, 15 Oct 2012 18:37:52 -0700

I had a great dinner with Anthony the other night and he remarked how I seemed a little happier. He wasn’t the first person to notice and I’ve been surprised at all the comments to that effect because I was pretty unaware. Before taking my vacation I was pretty sour on quite a few things.

I’ve been bootstrapping an education startup for a little over a year and felt that we lacked focus. Of course, when bootstrapping everything tends to move slower and near term successes can be misleading. They can also distract you from validating your business model, my primary concern. Not being a patient person also doesn’t help. But the main thing was that I hadn’t had a break from the grind of work and startup for a long time.

Other than posting pictures to Facebook of my travels, I was pretty unplugged and it allowed me a lot of time to reflect. I also had a close friend suggest that I needed to think a little bigger with my startup. Actually a few friends repeated this theme. The words didn’t really hit me until the end of my vacation and I had time to absorb them. If I hadn’t had the break, I’m not sure I would have been able to absorb them.

I have no problem working day after day after day. I feel guilty taking a break. I feel guilty not working. But this break showed me that in no uncertain terms not taking a real, effective break negatively coloured my judgement. It’s a little counter intuitive (for me) but to be better execute my startup, I need to take meaningful breaks now and then.

Getting started with GTD and The Secret Weapon

Sun, 14 Oct 2012 13:23:00 -0700

I had two great vacations in a row. First I went to Rotterdam, Amsterdam, and Barcelona where I got to visit the family Roos in Rotterdam and some Schulich alumni in Barcelona. It was a lot of fun as I got to catch up with good friends I hadn’t seen in a while and visit Barcelona for the first time. Rotterdam was great and Barcelona was fun but exhausting. I really enjoyed myself and there’s something I seem to identify with when I’m there. I could have easily moved to Copenhagen and Rotterdam is interesting too. Your castle tends to be smaller and you tend to have less stuff…but I find that very freeing. The week after returning we hiked the grand canyon. The trail is very well marked and fairly easy going. However, there’s no hiding the fact that you are losing (and gaining) around a mile of elevation. That combined with the heat requires ensuring you keep hydrated and salted less you end up with some pretty harsh cramps.

Two great trips back to back put me quite far behind at work and this was compounded by the fact that we launched an initiative right before I left. What saved me was implementing the basics of Getting Things Done for my inbox. By implementing GTD, I didn’t have to keep the lists in my head and was able to focus on the work. A side benefit is that I really felt that I was accomplishing things as I crossed them off of my Action Pending list. I tried implementing GTD in the past but I always failed because I didn’t have to. Now, I didn’t really have an option.

On Friday, I took it one step further and combined GTD with Evernote as detailed on The Secret Weapon. I did this because I still had multiple action lists. I had a list for my work at Microsoft, a list for my work at Alpenspruce, a list for my personal email, and a personal to-do list. The results have been very positive; I now have mostly one list. I say mostly because there’s always something that seems to slip out but that is now the exception rather than the rule.

The biggest benefit so far is that I am making more effective use of my time. If you lack natural organizational skills and want to make more effective use of your time, check them out!

Great video on using Mechanical Turk to help with Customer...

Tue, 07 Aug 2012 18:06:45 -0700

Great video on using Mechanical Turk to help with Customer Discovery by @danshiparo

Consulting is risky

Tue, 24 Jul 2012 23:07:00 -0700

Last Friday I went home for my dad’s 70th birthday. Before leaving for the drive, I met with my co-founder to sync up on a project and to get some work done on our lacking website. It’s one of those things that we just haven’t had time to work on as we’ve both been working full time jobs and building up our new business on weekends and in the evenings. We’ve taken what we both thought to be a very low risk approach: build the startup by taking on projects that pay. I recently came to the conclusion that it is actually a high risk strategy.

My view may appear a bit contrarian. What would be low risk about taking a project based approach? We work with our customers to improve and support their business processes and we get a check at the end. At the same time, we’re looking for a recurring problem that we can then productize…uggg I hate that word. paradigm shift, upward win-win spiral. I digress. It would appear that there is no risk as we get paid for everything we do and if we’re lucky to find a product opportunity then great! There are several problems with this approach.

The first problem is speed: we are too slow. Once we enter project mode, the business shuts down and we focus on completing the work. We’ve been fortunate to have great customers but as I have viewed this as a search for a product opportunity, I’ve put aside most of my normal project management rigor as a “cost of learning about the opportunity.” I don’t know our margins but I know they are razor thin. We serve great clients but they are under tremendous financial pressure and margins are already ridiculously thin. We cannot scale this as a professional services play because the margins just aren’t there so we must do the heavy lifting, which our clients love - they are traditionally underserved - but this hinders our ability to build a scalable business.

The project work also leads to false positives. Clients think our solution is great and will be “used by every school district” and we believe it too; we start to get excited and feel no need to validate the opportunity. For a recent project, I decided to follow Steve Blank’s advice and get out of the office. I made a few calls to test the value proposition of one of our “can’t miss” product opportunities. After a few calls, I was crestfallen. It turns out that while value exists, the total available market is too small. So, what’s the problem? Sure my feelings were hurt but I got paid! Cash in the bank! No risk! The problem is that it took me a few months to complete the project - remember, I’m working evenings and weekends - and only after did I test to see if it was a scalable opportunity. The problem is relatively simple and well defined so it could have been tested in a week and I could have had positive confirmation if this is a scalable opportunity.

The last problem I’ve come to appreciate is taking on projects delays finding a business model and forces you to adopt one: services. The services model is not a bad one and plenty of people have made a lot of money providing services. Just ask a partner at McKinsey. However, we serve the non-profit sector and there’s a reason most successful professional services organizations provide pro-bono work to non-profits. It is challenging to scale.

Now, if you haven’t read Search vs Execute go read it now. It’s probably the most important thing you’ll ever read on startups and this goes double for anyone with an MBA. If you have an MBA, you probably dismissed the link…no one could possibly tell you anything about business…you’re a master of the universe…so I’ll paraphrase: a startup is not a smaller version of an established business.

So, as we’ve been doing projects, we’ve pushed off our search of a scalable business model because we have to execute on projects for our customers. In following the services model, we’ve spent a lot of time discussing things like our website and how we’ll sell our services to customers.

What starts out as a low risk way to building a business - you’ve got customers paying for your labour - turns out to be very high risk.

I should mention that we’re not kids and we’ve got bills to pay. But if I were truly honest, I’d say we have lifestyles to pay for. We like our wine and our evenings out and our weekend excursions. We are very comfortable in our golden handcuffs and that’s the most insidious risk of all.

Tonight I registered a domain name and this week I start testing…

If you’re in or have been in a similar situation, I’m interested in your thoughts on the subject.

You don't scale

Sun, 15 Jul 2012 18:21:11 -0700

I recently overheard an experienced, educated consultant advocate to a client, “The nice thing about us is that we can do it all: sales, marketing, development, etc” and as the words came out of his lips, I was overtaken with a sense of foreboding. It brought back a memory of my dad saying of his once very successful business, “One of my biggest mistakes was trying to do it all…”

I think the feeling stemmed from years of working with half-assed engineers unable to deliver, employers unable to view a relationship beyond a single project, and clients that had difficulty articulating their needs. There is nothing worse than promising something to a client and then being unable to deliver. You constantly think, “the only way to do something right is to do it yourself.” Fortunately, new software development methodologies have come along in the past 10 years to help with most of these issues. Test-first engineers tend to deliver robust value quickly. There are exceptions, but chances are if the engineer can write a test, they’re more likely to have a much stronger grasp of the problem and hence the solution. With the new methods you can deliver value into the hands of the customer faster. It’s great to have QA…and you should have QA…but no amount of QA can make up for a busy customer that doesn’t read your great documentation. It’s been my experience that customers only say anything of value if they can put their hands on the product. By iterating, you lower (not eliminate) the amount of rework necessary because the customer was unable to articulate their needs clearly. Since you end up with less rework, you have employers more open to the idea of looking longer term.

I believe two additional motivations were not having a “boss” and being able to spend less time “working” and more time doing “fun” things. This is something I hear from lots of folks and something that I’ve wrestled with a bit over the past few weeks. For me, fun ultimately boils down to building a scalable business. That’s why I’m so impatient looking for a successful business model to help school districts. But I think this consultant may have a somewhat different view of fun.

I recall a professor suggesting that “not having a boss” as one of the worst reasons to start a business. You think you’re winning because you “don’t have a boss” and make great money. Of course, you don’t account for your salary and if you did their return on investment is dismal. You often see this with families that start convenience stores. They put in tons of hours and think they’re making a great return, but they don’t subtract their salary. People that take this approach often have tax minimization something that’s very important to them - “I’m not giving my money to the Man” - and it works out for them because their profit tends to be so low that they pay minimal taxes. Success!

“Doing it all yourself” is very difficult to scale and it’s really difficult to scale consulting unless you have a client with a very expensive problem you are uniquely positioned to fix. To me, doing it all yourself equates to buying yourself a job without the benefits of a job.

I noticed a few articles and tweets about the value of agile...

Sun, 15 Jul 2012 16:12:00 -0700

I noticed a few articles and tweets about the value of agile methods over the past couple of days. I can’t speak for everyone but by using BDD or TDD, I find myself landing on better design, having greater confidence promoting versions to production, and ensuring that a change in one area doesn’t negatively affect others.

Using Small Changes to Drive Enterprise Adoption

Sat, 14 Jul 2012 14:00:00 -0700

For the past two years I’ve helped drive adoption of a new enterprise application portfolio management process at Microsoft IT. A core concept in EAPM is information communication technology (assets). Needless to say, there are strong opinions about ICT at a software company and there are ridiculously strong opinions at a traditionally dominant software company. There has been no shortage of extremely bright, passionate people trying to foster adoption and our initial efforts mirrored their thoughts on what the minimum bar should be when describing ICT. On top of that, all of the conventional wisdom and experience reflects their positions. Pick up any book on managing information technology and you’d have a hard time arguing against their positions. Of course it doesn’t hurt/help that they’ve written many of the books. For example, the information about an application doesn’t become interesting or really useful until you start to describe its relations to other things within the enterprise. It makes sense that you require the owner to describe those relationships when they create an application record. Any reasonably good software architect would expect nothing less. We initially asked users to describe about 10 relationships when registering an application. Unfortunately, good software architects are rare and often extremely busy so the tasks were delegated to folks much less knowledgable about the software and this created substantial organizational resistance. It was a very frustrating experience because we thought we were asking for the bare minimum. It was also frustrating for these users who didn’t see the value in doing “all this work.”

Eventually, we figured out that our big change was cultural and that we had to get buy in so we took a different path and asked users to provide us the bare minimum amount of information. The ask was so small that no one could reasonably refuse the work without looking foolish, which removed the political roadblocks. Another win is that customers are now asking us to provide ways for them to describe the ICT relationships we initially thought were needed.

Lessons learned

What is viewed as a small change in one organization or group can potentially be a massive change across the enterprise
Your ability to affect change is very limited, consider the 15% principle when driving change
Enterprise wide process adoption is as much about changing culture as it is about the process
Users need to perceive value otherwise you’re going to need an executive mandate and even then you’ll end up with very low user satisfaction

Migrating from MongoDB to Postgresql

Sat, 14 Jul 2012 13:23:01 -0700

I also use BDD and continuous builds so that I can get valuable software into the hands of customers quickly. Nothing makes me more uncomfortable than the old manual “copy-paste” of untested code into production. On the downside, when customers get access to an application that makes their lives easier they tend to use it. I also like to learn new technologies and techniques and when I recently gave customers access to a “functioning prototype” they started using it. I didn’t have the heart to tell them they had to go back to pen and paper.

All the hipsters are using NoSQL because database schemas are no longer ironic. Not wanting to be left behind, I pulled on my skinny jeans, put on my fedora, and implemented a RoR based prototype using MongoDB. It was very interesting and the schema-less part was fun. No more lame-o migrations. After quickly putting out the fully functional prototype - I love Rails - my customer started using the tool and was very excited. I didn’t think much about it until they started asking for reports (aggregations). I thought, “Not a problem just aggregate these documents…” No aggregations. Of course, you can get into map/reduce and all that good stuff but I have a day job and a wife. I didn’t have time to do something that would take me seconds to do in SQL. So I pulled off my skinny jeans and handed my fedora back in and reached for Pentaho Kettle, which fortunately has a MongoDB connector. Over the course of a few evenings I was able to migrate my data into my new schema and because all of my code was tested, I was able to switch to an entirely new backend with a very high degree of confidence.

I see a place for NoSQL solutions like MongoDB but RDBMS like Postgresql are damn good at what they do.

Lessons learned

NoSQL is interesting and solves some interesting problems but so does RDBMS
Despite being able to deploy multiple, fully tested, releases a day into customers hands, sometimes you need to say “Wait.”
Skinny jeans are itchy