Six-Month Retrospective of a Solo Side Project (Revenue? A Game? No, It Was Fun)
After work, I made a weather alert app even though I knew nobody would use it. It's called 'Today's Weather Ttukbaegi.' At first it was just a simple widget that showed the weather with a sky image, but I added a feature that automatically pops up subway congestion alerts around commute times.
The technical challenges weren't actually that big—just hooking up a map API and adding FCM push notifications. Still, as a solo developer, the hardest part was 'constantly revising my own plan.'
Here's what I learned along the way:
- Even 30 minutes a day, at the same time every day, makes the difference between a polished product and one that isn't.
- Keep cutting the scope—an MVP really should include only the absolute minimum features.
- If the first review after launch is praise rather than a threat, that alone is touching.
- To get real user feedback, you have to mobilize communities, acquaintances, and social media.
In the end, the monthly active user count has always been in the single digits, but the biggest joy came from seeing code I wrote outside of work actually running on my phone. Now I'm slowly working on a second version rebuilt in TypeScript, and I guess for developers, the act of building itself is the fun.
8 answers
Yeah, fun's the best lol
If it's single digits, it's practically nonexistent lol. Still, if it's fun, that's all that matters.
My side-project app also has zero users besides acquaintances... but still, because it's something I actually need and use, it makes me code for 30 minutes every day. 6 months is a pretty long run.
I agree that narrowing the scope is the right call, but then why add the commute congestion alerts? lol. Still, the most important thing is not obsessing over user numbers and being satisfied with what you do.
Well, the weather app market is already fully dominated by Naver and Kakao, so there's no clear differentiation. Moving from a sky-image widget to congestion alerts is a good direction, but haven't they failed to validate whether that's actually what users want? It doesn't quite align with the whole MVP talk.
How do you get the congestion data? Just curious.
I also tried making a timetable widget myself, but no one used it, so I dropped it. Still, what I learned from that process really helped me later at work. The fact that you kept at it for six months is impressive by itself, and if you gained that kind of enjoyment rather than results, you've already succeeded.
That's right lol, the answer is doing 30 minutes a day consistently.