I’ve been writing software for over a decade. I know how to build things. I know how to architect systems that scale, handle failure, and hold up under production load. I’ve shipped code in regulated industries where a bug in the wrong place has real consequences.
None of that prepared me for figuring out how to get strangers to install my app.
When Permly was ready to launch, I realised I had no mental model for what came next.
My whole background — my studies, every job I’ve had — is technical. Product features come from requirements. Requirements come from stakeholders or users. But when you’re the one starting from zero, there are no stakeholders and no users yet. You have to go find them.
I didn’t know how to launch a product. I didn’t know what distribution channels made sense for an indie Android app. I didn’t know what a reasonable CAC looked like, or how to write copy that converts, or which communities were worth posting in versus which would ignore you.
I posted on Reddit and Indie Hackers. That was it for launch day. No coordinated campaign, no pre-launch email list, no concurrent posts across multiple platforms. Just two posts and a link.
Looking back, that was the main thing I’d change. The going-live moment is not a joke. If you’re going to make noise, make it all at once — social, communities, newsletters, whatever you have access to — so the signal compounds instead of trickling in one post at a time.
I’ve been running Google Ads since launch. Around €350–400 a month. A few thousand installs so far.
Zero revenue from that.
Paid volume is useful for one thing: it gets you real users fast enough to start getting real feedback. That’s worth something. But as a revenue strategy it’s not looking promising, at least at my current scale and without a more sophisticated conversion funnel. The unit economics don’t work unless you’re converting at a rate I’m not hitting.
What’s actually working is direct. Word of mouth from the Reddit and Indie Hackers posts. Personal conversations. Sharing it with people I knew had the problem. The installs from that are smaller in volume but producing actual feedback — users who found it themselves, who understand why they’re using it, and who have opinions about what they want next.
The clearest signal so far: people want a snooze mode. Block notifications from this app for the next [x minutes]. Not a permanent profile switch — a temporary, timed block. That’s a real use case I hadn’t fully anticipated, and it came from people who actually installed the app and tried to use it in their daily life.
You don’t get that from an ad impression.
The strategy from here is slower but more sustainable. New features shaped by user feedback. Broader distribution through platforms I haven’t posted on yet — Product Hunt, Hacker News. Organic search traffic through the website.
None of these are fast.
I’ve spent enough time on the other side of this coin — building features that felt ready to ship but sitting on them because something else needed to be perfect first. I’ve seen that pattern kill projects that had real potential. With Permly I decided to be different: ship the thing, let real users break it, and learn from what actually hurts.
That’s not a comfortable way to work. But the alternative — waiting until you feel ready — is how you end up with software that’s technically finished and has no users.
I have users. Some of them are paying. Most of them are not, and that’s fine for now. The product exists, it works, it solves a problem that is real, and every piece of feedback that comes in makes the next version better.
That’s the job. The marketing part I’m still figuring out.
Permly is on the Play Store if you want to try it. And if you’re building something solo and want to compare notes, reach out.