What I learned watching medical students build apps for their own class

I spend a lot of time around medical students (I am one), and lately I have watched classmates quietly build tiny apps for our year group — a quiz from this week's lectures, flashcards, a tracker for clinical rotations. Small things, used by maybe 30 people.

A few things surprised me:

1. Nobody wanted an audience. Every app was made for a specific group chat. The moment I asked "would you put this in an app store?", the answer was always no — they just wanted the link to work when they dropped it in Telegram.

2. The first version took minutes; the fixes took the real time. Someone would complain in the chat that evening ("the quiz shows the answer too early"), and the maker would fix it the same night. The feedback loop was the product, in a way.

3. Shared data mattered more than features. The apps that stuck were the ones with a simple shared board — who has finished this week's set, a class leaderboard. Solo tools got forgotten in a week.

Full disclosure: I work on Koi Launch, which is what some of them used. But I think the behaviour matters more than the tool — people seem to want software at the scale of a group chat, not the scale of a market.

For those of you who have built something for a small group — a team, a class, a family — what made it stick, and what made people quietly stop opening it?