Yesterday I received my daily deluge of campaign fundraising emails, one of which reminded me that it was the last day for early voting. I knew I’d be busy on Tuesday, and one of the early voting locations was near my gym, so I decided to cast my ballot before my afternoon run. Upon arriving, shocked at the number of Chicagoans who also thought voting on a Sunday was a top priority, I took my place at the end of a line that extended out the door with several switchbacks in between. During my wait, I experienced three distinct events that reminded me of key lessons I’ve learned as an entrepreneur over the past five years building GQueues, the leading Google-integrated task management service for people and teams.
A few minutes after taking my place in line, volunteers appeared offering coffee and donut holes. Most of us happily accepted the thoughtful gesture. People were cheerful and congenial on this mild fall day. But those of us at the end of the line were anxious; although our appetites were satisfied, our greatest desire was still unmet. We wanted to know how long we would be waiting in line, of course! Taking the lead for the group, I approached a person about halfway between the end of the line and the door and asked how long he had been waiting since standing in my position. 30 minutes. Great! We now knew we’d be voting in approximately an hour. Though the volunteers were well-meaning, they overlooked one of the most important services they could provide -- sharing ballpark wait times -- so people could decide whether if it would be better to vote today or wait until Tuesday.
Almost all the features I’ve built for the online task manager GQueues have come directly from customer feedback. However, the coffee and donuts situation reminded me of a few times when I’ve made changes to GQueues without getting user input first (it turns out people don’t like having task colors changed on them overnight, and they don’t want PayPal as the only payment option)! The Lean Startup movement emphasizes the idea of getting out of the building and talking to customers to ensure you are building the right solution. As entrepreneurs, we are always thinking about how we can improve our products. We have theories on what will solve customers’ problems and improve the business. Our vision of the future is essential to our success. However, our instincts and plans must always be rooted in a clear understanding of users and their needs -- which only emerges through listening. Asking customers for feedback with open-ended questions and then actually listening is paramount to building a truly valuable service. This isn’t revelatory. Most successful startups embrace this concept. But my time waiting to vote brought this to life again for me. Without truly listening, we may end up giving customers what we think they want (coffee and donuts), without addressing their most critical needs (wait times)!
One of the aspects I enjoy about voting in person is the feeling of community at the polling place. While waiting in line, I listened to a Chicago Teachers Union member and a Chicago Transit worker discuss their views on the current state of the city and their support for various politicians. After an hour we made it to front door of the building, just as calculated. With a sigh of relief, we walked inside, only to see the line of people snaking down hallways throughout the building. I realized my error in reasoning. After talking to those at the front of the line, we learned it was another hour-long wait once inside the building.
At that moment, I recalled a key concept I had learned about project planning in my Masters of software engineering program: always identify the areas of highest risk -- the greatest unknowns -- and address those first. Focusing on what you don’t know is even more important than what you do know when embarking on something new. When building the GQueues Android app my biggest “unknown” was whether I could create a relational database in Android that could sync with the object-oriented “NoSQL” datastore in Google App Engine. If I couldn’t get syncing to work between the two types of databases, there was no point in building the app. So before I started any actual development, I spent an entire week exploring and testing this one area to answer this question and eliminate the risk before moving forward.
As early entrepreneurs there is a huge amount of information we don’t know -- about our market, our potential customers, our competitors, and the future in general. Learning as much as possible in these areas is critical to discovering a viable business model. And yet we often must make decisions with incomplete information. Consequently, it’s crucial to recognize what we don’t know when making decisions, so false assumptions are avoided and risks can be mitigated as much as possible. Had I focused on what I didn’t know at the polling place, I would have either sought more information or at least considered the impact of the unknowns in my decision to stay or go.
Standing inside the building, I realized I should re-evaluate my situation. I’m sure many people stayed in line because they really couldn’t vote on Tuesday, but I was in a position to compare my options.
While considering my choice, I recalled the Sunk Costs Fallacy I had learned so well when I made a major shift in GQueues’ mobile strategy. When we make a decision based on previously incurred expenses, instead of the cost and benefits that an action will have in the future, we are falling subject to the sunk costs fallacy. In 2011 I built an HTML5 mobile web app instead of separate native apps for iOS and Android with the idea that I could provide offline functionality to mobile users with only one code-base to maintain. The mobile web app was welcomed at launch, but I spent the next year trying to address user performance issues associated with not building native apps. Although I had invested three months developing the mobile web app and countless hours trying to improve it, I realized this was irrelevant to considering what was best for GQueues customers. I abandoned the mobile web app and decided to finally build native Android and iOS apps, admitting my mistake publicly and moving forward.
In business and any area of life, it’s easy to fall into the trap of the sunk costs fallacy, because we get attached to our investments of time and money. Recognizing the faulty logic though can help us avoid making decisions that don’t lead us into a brighter future.
Although I knew I’d be busy on Tuesday, my schedule was flexible, and I was willing to bet that if I went during the day while most people were at work, the line to vote would be less than an hour. I said farewell to my fellow voters, passed along the expected wait time to those at the end of the line and headed for the gym.
UPDATE Tues, Nov 4, 2:20pm: I voted today. Perfect timing. No line. In and out in 8 minutes flat!
I love building products! And Python. And dark chocolate. When I'm not leading the team at GQueues, I can be found running ultras on the trails of the Rocky Mountains.