So a while ago I built a small app for my wife. She does this thing where she buys stuff cheap at car boots and markets and flips it on eBay and Facebook, and her whole system was the notes app and her memory. I made her something to track it — FlipperHelper — and then people on Reddit actually wanted it too, so I put it on the App Store. Free, no backend, no ads, and honestly no real plan.
It's been about four months now. Around 360 people have downloaded it, roughly 140 of them actually use it, the website grows its traffic every week, and there's an Android beta now. None of that came from money, and I did a lot of it wrong before I did it right. So this is just me writing down what actually worked, in case it's useful to someone doing the same thing.
The first decision that mattered was building it with no backend, on purpose. Every project I made before this one I over-engineered for a flood of users that never showed up. I'd pay every month for servers "in case it takes off", and that fear made everything slower and heavier. This time I did the opposite — everything runs on the phone and syncs to the user's own Google Drive when there's internet. If a thousand people install it tomorrow my costs stay zero and nothing falls over. And the funny thing is that turned into an actual feature, because flea markets often have no signal, so people know their data is always there.
Reddit got me my first users, which I'm grateful for, but it hits a ceiling pretty fast. The niche reselling subs only have so many people, and after that anything you post starts to read like an ad no matter how much personal story you wrap around it. To be honest I never learned to write a post that doesn't feel like advertising to them, and these communities have seen a hundred people try to "add value" on the way to a pitch, so they're skeptical, and fair enough. It worked, but I couldn't keep pulling the same lever.
So I made a website and started writing for it, and this is the first time in my life I got traffic that grows on its own. The way I do it is not clever — I ask ChatGPT for keyword ideas, check in Google Keyword Planner which ones people actually search, and then for the good ones I just record a long voice note of everything I know about that topic and turn it into an article, then clean it up for search. It's slow. It takes a month, sometimes half a year, but it compounds and it finds you. Right now Google is about a fifth of my site traffic and — this one really surprised me — ChatGPT is around another tenth, already bigger than Reddit and every launch directory put together. I tried Instagram and TikTok too but for me it was too much work for nothing that lasts. Two days writing articles gives me traffic for a year. Two days on a video nobody watches and it's dead the next morning. So I'm not doing short video right now.
The other surprise was the App Store itself. People just search it, find the app, and install it, every single day. It's not the 10,000-downloads-a-day thing you see people posting, it's a slow trickle that went from about one a day at launch to four or five now, but it's free and it keeps climbing.
Now the mistake. I shipped with no analytics at all, and I was weirdly proud of it — look at me, the privacy guy who doesn't track anyone. What actually happened is that the App Store only shows you data for the small share of users who opt in, so at my size I was completely blind. I had no idea how many real users I had or if anyone came back. I gave in and added tracking for just two things — when someone adds an item and when someone marks it sold — and within two weeks I could finally see the truth. About 40% of people come back the next day and roughly a third go all the way from adding to selling, which is the whole point of the app. You can't fix what you can't see, and I was guessing for two months. Add analytics on day one. You don't need to spy, just measure whether the thing worked.
The same lesson hit me with feedback. With no backend there was literally nowhere for people to reach me, so I added a "find us on Reddit/Instagram" line, and it half worked, except I kept noticing people creating brand new accounts just to message me. That's a bad sign — if reaching you means signing up somewhere, most people just won't. I copied a feature I saw in another app: a button that opens your own email with everything already filled in. Everyone has email. The first person who used it sent me screens and screens of feedback, three features I've since shipped and a pile of bugs, and it's shaped nearly every update since.
I also finally started asking for reviews properly. I ask right after the app actually paid off for someone — the moment they mark an item sold — and instead of asking once and giving up, I ask again every week or so, because Apple lets you ask a few times a year. That caught my core users who'd tapped "not now" the first time while they were busy. I only have nine ratings, but about half came from that one change, and I felt a real bump in installs and visibility after they landed. In a small niche each rating is worth a lot.
Now I'm building the Android version, because the AI assistants that recommend the app keep pointing out it's iPhone only, Android people already land on my site and leave because I have nothing for them, and the demand in the closed test is real. It probably doubles my installs. It can't be one of the cross-platform wrappers though — the app is offline first and I need real device access and to keep it fast even with a hundred thousand items in it, so I'm just building it twice, still with no backend.
If there's a point to all this, it's that I'm in no rush. It's a passion project, I have no revenue target, and that's exactly why I can keep listening to people and building the thing they actually ask for instead of launching something raw and pouring money on top. Patience and free channels and actually reading your feedback beats a big paid launch, at least it did for me. Anyway — that's where I'm at.
Top comments (0)