How to launch an app when nobody knows you exist.

The launch you're imagining — the Product Hunt badge, the write-up, the hockey stick — quietly assumes someone is waiting for your app. For a first app from an unknown developer, nobody is. That's not a reason to skip launching; it's a reason to run a different play. Here's the sequence that works from zero.

Every channel is a spike. Plan for the decay.

Start with the shape of the problem. Every external channel — a Reddit post, a newsletter mention, a good day on Hacker News — produces the same curve: a burst of visitors, then a decay back to baseline within days. Channels don't compound. Only two things do: store search, which grows as your listing earns rankings and ratings, and retained users, who come back and bring other people with them.

So a launch for an unknown app is really a sequencing problem. Build the compounding assets first — a product that keeps its first users, a listing that converts strangers — then spend your spikes into them. Spend a spike before those assets are ready and you've burned a channel, and most of the good ones only fire once. You can't post the same "I made a thing" story to the same community twice.

Week one: release, and tell no one

Ship the app before you announce anything. The first version strangers meet will have problems you can't see from inside: a crash on a device you never tested, a first screen that confuses people who didn't watch you build it, a permission prompt at exactly the wrong moment. You want those found by a trickle of quiet arrivals, not by the biggest audience you'll ever assemble — because early ratings sit on your listing long after the bugs are fixed, and every visitor your launch spike brings will read them.

Google Play forces this discipline on new personal accounts anyway: twelve testers for fourteen unbroken days before you can touch production. Apple forces nothing, so impose it on yourself — TestFlight with whoever you can find, then a quiet release, then a week of watching analytics before you spend any attention. The number to watch isn't installs, it's whether the people who install today open the app again this week. If they don't, you have a first-session problem, and no launch channel fixes that. It just exposes it to more people.

Week two: the listing is the multiplier

Every channel you ever fire dumps strangers onto the same product page. That page's conversion rate multiplies every spike you'll ever get — a listing that converts twice as well makes every future Reddit post, newsletter mention, and search impression worth twice as much. It is the single highest-leverage thing an unknown developer controls completely.

For most first listings the gap is concentrated in two places: the screenshots — only the first two or three render in search results, and raw UI dumps with no headline are the default mistake — and the name-plus-subtitle pair, which is the only indexed text searchers actually read. Spend the week here. This is the one place a solo developer should aim for a genuinely professional standard before launch, because it's the part of the funnel that never sleeps: it's still converting some stranger at 2am, eight months from now, long after the launch week is forgotten.

Weeks three and four: fire one channel at a time

The instinct is a coordinated blitz — everything everywhere on launch day. For someone with an audience, that concentration makes sense. For someone without one, it's backwards: your first channel will teach you things that make the second channel land better, but only if you leave a gap between them to act on what you learn.

Start with communities you already belong to, and post as a member, not a marketer. "I spent eight months building this because X annoyed me, here's what I learned" earns a conversation; a press release earns a removal for self-promotion. Answer every comment. What people misunderstand in the thread is a preview of what store visitors misunderstand silently — feed it back into the listing before the next shot.

Then the lottery tickets: Show HN, Product Hunt. Treat them as exactly that. The expected value is real, the variance is enormous, and either way the traffic is gone in a day or two — so take the ticket, but don't build the plan around winning it. Slower and more reliable: the niche newsletters and YouTubers who cover your specific category. A mention in front of three thousand people who care about your problem beats a cold email that a general tech journalist never opens. And nominate yourself for featuring — App Store Connect has a featuring-nominations section where you pitch your own app to Apple's editorial team. No guarantees, but the polished-listing-from-an-unknown-developer story is precisely the kind of thing editorial teams exist to surface. Play featuring has no equivalent form — Google's editors pick, so there the listing itself is your only pitch.

One more tool if you have even a small trickle of pre-launch interest: both stores let you bank it. Apple pre-orders let people commit up to 180 days before release and auto-deliver on day one; Play has pre-registration. Scattered interest collected over weeks becomes a concentrated day-one bump — the only spike whose timing you fully control.

Measure the spike, not the spike's size

Raw visitor counts are the least useful thing a launch produces. Two thousand visitors who don't install is a listing problem; two hundred installs that don't come back is a product problem — and each spike tells you which one you have, for free. Both consoles break acquisition down by source, so between channels, check two numbers: how the spike's page conversion compared to baseline, and how that cohort's day-one and day-seven retention looks. Fix whichever is broken before firing the next channel. The unknown developer's one structural advantage is that most channels are still ahead of you — the only way to waste it is to spend them all into a broken funnel during one loud week.

Do the week-two work in an afternoon

The listing pass is the step most launches skip because it feels like a design project. ShotCanvas collapses it: drop in raw screens, pick a template or let AI compose the set and write the headlines, and export pixel-exact screenshots for every App Store and Google Play slot — so the multiplier is in place before you spend a single spike.