I built a driving test app. Getting anyone to see it was the harder part.

Building the driving app was the fun part. I already told that story, spaced repetition, a real problem, an idea i could not let go of. This post is the less fun half: what happened after i shipped it and had to convince actual humans to open it.

Turns out building something and getting someone to see it are two completely different jobs, and nobody warns you the second one is harder, ngl.

I tried Reddit first, because it felt easy: write a post, drop a link, done. It drove exactly zero installs i could trace. Not "not many". Zero. I sat there refreshing analytics like the numbers might change if i stared hard enough. They did not.

TikTok was the opposite story. One video landed in front of almost a thousand people through the For You page and pulled in new followers i did not have the day before. Same effort, wildly different result. So now i know: for this app, for this audience, TikTok is where the people actually are, and Reddit is not, at least not the way i tried it.

Then came the part where i properly embarrassed myself. Twice.

First time: i pull up my analytics, and it looks like checkout is completely broken. Fifteen failed purchases, not one success. My stomach drops a little, because a broken checkout means nobody can ever pay me, ever. So i go digging, person by person, email by email, and every single failure traces back to me. My own test accounts, my own dev builds, one from the Play reviewer. Real users had not even tried to buy anything yet. I built a whole crisis out of my own noise.

You would think i'd learn. I did not, not immediately anyway. Weeks later i am looking at the onboarding funnel and it says most people who start onboarding never finish it. Bad number, right? Except when i actually pulled the list of who dropped off, most of them were, again, me, plus one extremely persistent robot: Google Play runs an automated pre-launch test on every build, and that bot was sitting in my funnel data too, cheerfully failing to finish onboarding like a very bad user. Once i pulled the real humans out, the small handful of them who tried it mostly made it through fine.

So the actual lesson, twice learned the hard way: get your own noise out of the data before you trust what it is telling you. It is astonishingly easy to scare yourself with a graph that is secretly just you.

And somewhere in the middle of all that, one honest signal did show up. A real person, in Uganda, on a real phone, made it all the way to the actual payment screen, Google's own checkout sheet, and then backed out. Not a bug. Not an error. Just a person looking at a price and deciding not yet. That is not the win i was hoping for, but it is the first proof the whole thing actually works end to end for someone who is not me.

Where that leaves things: small real growth, mostly from TikTok, and zero real payments so far. Not exactly the highlight reel version of a launch story. But two lessons i am carrying into whatever i build next: trust the data less than you want to, especially your own, and distribution matters more than i gave it credit for. A good post or a good feature with nobody behind it teaches you nothing, because there is no one there to prove you right or wrong.

Still going. If you or someone you know is prepping for the Uganda theory test, have a look, and if you happen to be the one person in Uganda who almost bought it, feel free to tell me what made you stop 😅

Thank you for reading. Tschüss 🤓