Today I shut down Susana.
Susana was a Slack bot for mandatory time tracking and vacation management in Spanish companies. I started it in February 2025. Twenty-one months, ~33k lines of code and 328 commits later, I'm closing it in an orderly way: teams got their data exported, there are no subscriptions left hanging and no leftover costs.
The short version: the product worked, but the traction wasn't enough to justify keeping it alive.
In Lose your first 50 games as fast as possible I wrote that every lost game teaches you a shape. These are the two shapes I'm taking from this one.
1. I built far more than the traction justified
Susana ended up with a manager hierarchy, a permissions system, a full onboarding, a blog with a CMS, several pricing plans and everything needed for the Slack Marketplace. All of it worked. And a lot of it was built before the market had told me whether it wanted any of it.
Building is the comfortable part. When traction is slow, writing another feature feels like progress. It isn't. Every week spent on the product was a week not spent getting it in front of the right people.
Lesson: put the smallest core in front of real customers first, and let traction decide what gets built next. Development should follow demand, not the other way around.
2. Always charge. No free tier.
I was generous with the free tier. Small teams could use Susana for free, forever. It felt like the friendly thing to do, and it looked smart against competitors who charge per seat.
It was a mistake. A free plan makes it too easy to say yes. Nobody had to stop and decide whether Susana was worth paying for, so I never got a real answer to the only question that mattered. Usage without payment feels like validation, but it isn't.
I removed the free plan in October 2026. By then the habit was set.
Lesson: charge from day one, even if it's little. One customer paying €50 tells you more than ten using it for free.
A few more things I'd do differently
- Decide pricing by talking to customers, not by rewriting it in code. Mine changed three times.
- Keep a steady rhythm. Side projects that move in bursts lose momentum, and distribution suffers most.
- Mandatory doesn't mean people buy. Time tracking is required by law in Spain, but most small companies already solve it with their accountant, a spreadsheet or a bigger HR tool.
What I'm keeping
The code base was solid to the end: no TypeScript errors, tests on the critical logic, real-time dashboards. Slack integration, payments and the time-tracking logic are pieces I'll reuse.
And the bigger thing: I set out to learn how to take a product from idea to market completely alone, and I did. Code, landing, pricing, payments, SEO, sales, support and, today, a clean shutdown.
The business didn't work. The lesson did. On to the next game.