The thing I built that nobody used

I once spent six weeks building something nobody used.

It was not a bad build. It worked. It did exactly what the spec said it would do. The problem was that I wrote the spec by imagining the person who would use it instead of asking one.

I found out the way you always find out. I shipped it, waited, and watched the usage sit at almost nothing for a month before I admitted what that meant.

The lesson was not “talk to your users.” Everyone says that and nobody argues with it. The real lesson was narrower and a lot more uncomfortable. I had preferred my version of the user to the actual one, because my version agreed with me and the actual one would have made me rewrite the plan.

That preference is the most expensive habit I have had to unlearn. It still shows up every time a project feels obvious.

Originally posted on LinkedIn.