Lessons Learned Trying to Build a Sustainable Software Business
For the past year I’ve tried to build a sustainable software product business. I experienced more failure in this season than any point in my life. I also stumbled across a few things that worked out well. In an attempt to not waste any of it, I’d like to document a few of them here.
What didn’t work
Any and all attempts at marketing
Trying to talk to my target market was like shouting into the void. Nothing I could do or say was compelling. Being compelling is partially about what you’re doing and partially about how you talk about it. I failed catastrophically on both fronts.
If I do something like this again I’d prefer to have a cofounder. Someone to debrief with. Someone to bounce ideas off of. Someone with their own scope of work. It changes the dynamic of getting feedback from “who have I not bothered recently about my project” to “this person wants this to be the best thing possible so I should let them see what I just made”. Or I could have formed an advisory board and given them each a small ownership stake.
Hard product pushes
At one point I took a 3 day work-cation this spring to help wrap up a big chunk of work. We have a small rv, so I drove it to a state park 45 minutes away with the plan of making some big progress. Between the planning, booking, driving, and setting up it wasn’t any better use of my time. There are diminishing returns on additional hours past a normal workday. I might have been slightly more productive if it was a longer trip but even then the returns would have been marginal.
What was a mixed bag
Prioritizing my learning
I partially chose my product based on the thought that “well if this doesn’t pan out at least I’ll have experience with technology x, y and z.” This turned out to be true. I became a competent Elm developer. I have more experience with WebRTC than anyone I know. This might turn out to be valuable, but it did not help me iterate as fast as possible to find product market fit. Deprioritizing my technical learning would have increased my chance of success.
Things that went well
Making a Podcast
Making a podcast was the most positive thing I did. I got lots of good feedback, got to have interesting conversations with people that I like and admire. It was a big investment in time though (~30 hours per episode). If I were to do it again I’d livestream the recording and then publish the audio unedited.
Regular contact with friends
I had a few friends that I’d video chat with on a semi regular basis that helped me keep sane.
Publishing a daily standup
For the last few months of the project I published a short daily standup about what I was working on and what I’d been thinking about. These almost never got more than a few views. It did help me prioritize my day inside of the void of outside feedback. It also helped me keep myself accountable to what I’d said I would do in previous days. I would probably have stopped sharing these to twitter earlier than I did, because (almost) no one was interested in them.
Increased flexibility for family
My oldest son started kindergarten this year at a school a mile from our house. I got to ride bikes with him there and back many times a week. I could have arranged doing this at my last job, but the freedom to not have to coordinate was huge. We had our third child this past January. I was able to take a good deal of time off at the beginning but then very slowly ramp up my workload.
The first few months I did some pretty extreme time tracking. I accounted for every minute of the day (sleeping and awake). This provided accountability and helped identify time I was working but wasn’t being effective.
I started out strong with retrospectives and by the end they were fewer and farther between. There was a set of questions that I would run though and update each week. They were:
- Daily number of hours tracked
- Daily number of productive hours
- How am I wasting time?
- How am I being productive?
- How do I feel about the week?
- Who was helpful?
- What needs to change?
- What am I afraid of?
- Who should I be asking for help?
- How well did I treat my [each member of my family]? (1..10)
- What is the 20% that is producing 80% of my results?
- What things could I automate?
- What will I accomplish next week?
Spending money to solve problems
Solving problems by buying stuff was always the most productive use of resources. Homepage templates, artwork, email templates all saved months of work for very little cost.
Quitting mid year
Paying taxes for half the year expecting to make 2x more creates a nice tax return that gave us some extra padding.
Maintaining a burndown chart
I tracked expenses monthly and projected those averages out into the future. It was helpful in making decisions about what to focus on as well as helped keep our family spending in check. We did well cutting back our spending while still being able to do lots fun stuff together.
Maintaining normal working hours
Keeping normal work hours was great. It helped me keep family obligations, gave my subconscious a chance to work problems, and helped my family respect of my work time.