Marketing for apps
What actually gets an independent app used after launch week — store search, retention, the communities where your users already are, and what to ignore.
Most advice about marketing an app is written for teams with a budget, and it does not survive contact with one person shipping something in evenings. This is the version that does: what works when the budget is zero and the time is a few hours a week.
The short version is that app marketing is mostly three things — the store listing, retention, and being present where your users already are — and everything else is a distant fourth.
Store search is your biggest channel, and it is free
Most installs of most independent apps come from somebody searching in the store. That makes the store listing a marketing channel rather than a formality.
- Put the search term in the title. Not only your brand. Thirty characters on the App Store, fifty on Play, and those words carry more weight than anything else on the page.
- The App Store keyword field is separate from everything users see. A hundred characters, comma separated, no spaces, no repeats of words already in your title. On Play there is no keyword field — the description is indexed instead, so it has to read naturally and contain the words people type.
- Screenshots decide the install, not the description. People look at two and leave. Caption them, and lead with what the app does rather than a sign-up screen.
- Ratings are ranking input. Ask inside the app, at the moment somebody completes something rather than at launch.
- Localise the listing before you localise the app. Translating the store page is an afternoon and opens store search in that language; translating the app is a project.
Then check what happened. Both stores report which search terms brought installs, and that report is the only honest source of what people call the thing you built.
Retention is a marketing channel
This sounds like a product concern and it is a distribution one. Both stores rank on engagement, so an app people keep using gets shown to more people. An app with good day-7 retention and no marketing usually beats an app with bad retention and a launch campaign.
Fix the first session before spending an hour on anything below. The reliable wins: no sign-up wall before the app has done something useful once, an obvious first action, and no empty state that requires setup to be interesting.
Be present where the problem is discussed
Not where founders are. Where your users are, doing the thing your app is about.
- Find the two or three places — a subreddit, a Discord, a forum, a Facebook group, a hashtag — where people discuss the activity your app serves.
- Read for a week. Answer questions without linking anything.
- Then mention the app when it is a genuine answer to a specific question.
This is slow and it is the highest-converting channel most independent apps have, because the people are pre-qualified and the recommendation reads as a person rather than an ad.
The things worth doing once, that keep working
- A landing page that is not the store. Something search can index, links can point at, and a person on a laptop can read.
- Directory and listing sites. Ten minutes each, permanent, and they accumulate. The list is in where to submit your startup.
- A short demo video. One take, thirty seconds, no music, showing the thing working. It goes on the landing page, in the store listing, and in every post.
- Make your site readable by assistants. "What app should I use for X" is increasingly asked of a model rather than typed into a search box, and the answer comes from what those models can read. A page that says plainly what the app does, in text rather than in an image, is most of the work.
What to ignore
- Paid installs, unless you know your revenue per user. Bought installs are the worst-retaining users available, and retention is what the ranking uses.
- Press releases and paid "featured on" placements. Nobody reads them and the links are usually marked
nofollow. - Posting the same launch announcement across ten platforms. It reads as what it is.
- Rebuilding the app because marketing did not work. Check retention first. If people who arrive keep using it, the problem is distribution; if they do not, no amount of distribution helps.
A realistic weekly rhythm
Two or three hours, spent like this: one hour in the communities answering things, one hour on the listing or the landing page, and whatever is left on a single directory submission or one update to the store screenshots. Repeat for three months before deciding whether it works.
The compounding is real but it is slow, and the most common failure is stopping at week six.
Where this site fits
WhatAreYouBuilding.AI is one of the permanent listings worth having: a page for your app, a regular do-follow link, and a place in a catalogue organised by country, category and funding stage rather than by launch date.
It is also queryable by agents via MCP, which is the part that matters for the assistant question above. You can check how readable your own site is with the free AI readability checker.
Free, and a person reads every submission. Submit a product, or see what a listing gets you.