How do you get your first app users? By hand, one at a time, from a list of names you can write down tonight.
The first users for an app come from people you can name, asked one at a time, and that is slower and less comfortable than building the app was. Most advice on how to promote a new app assumes an audience it never mentions, so it sends you to channels that need ratings, a tested claim or a budget you cannot judge yet. This page is about the first ten and the first hundred, not about getting more downloads once something already works.
Two kinds of reader ask this question and the honest answer splits between them. What follows: which one you are, what a list of forty names really produces, the twelve testers Google Play requires before your app can go public on Android, and why the install link is a bigger obstacle than the words around it.
See what a list of names producesThe short version
Nothing that gets the first ten users scales and nothing that scales gets you the first ten.
Asking people directly has no reach and a high hit rate. Store search, ads and posts have reach and, at this stage, almost no hit rate, because they need a listing, ratings and a claim that survives a stranger reading it. So do the unscalable thing first, on purpose, and stop treating it as the warm-up.
Be ready for the part that growth posts leave out. This is harder than building was. Building had a finish line you could see, and a bug either went away or it did not. Forty messages produce a handful of installs, most of whom open the app once. That is a normal result, not a verdict on your app, and the only thing that fixes it is another round with what you learned.
Two readers ask this, and they get different answers
The first reader built something for a group they already belong to: a five-a-side league, a studio, a class, a team at work. Your first users are a list and the work is administrative. Write the names down. Message them one at a time rather than in a group, say in one sentence what the app does, and ask for one specific thing. Expect to send follow-ups, because people who said yes forget.
The second reader built for strangers: parents, landlords, nurses, small shop owners, people whose day they have read about but never sat in. With no audience there is no shortcut worth your money yet. The work is to find a room where those people already talk, read it for a week before posting anything, then answer questions about the problem instead of pitching the app. That is weeks of effort, not a launch day, and it is the only channel that costs nothing but time.
Both readers should notice what is missing from this page. There is no list of thirty tactics, because reading a list is what people do instead of asking someone. Press, influencers, a launch post and an ad budget all assume something you do not have: an audience, a price you can afford per user, or a pitch that has already survived contact with a stranger.
What a list of forty names actually produces
Do the arithmetic before you start, because the drop at each step is what surprises people. Not everyone replies. Not everyone who replies installs. Not everyone who installs opens it twice. Four steps at a coin flip each leave you a small fraction of the list, and forty names becomes two or three real users.
The calculator below runs on your guesses. We have not verified a typical reply or install rate for a one-to-one ask, and any figure quoted as an industry average for this would be invented. Put in what you expect from your own list, then read the second number: how many people you have to ask to keep twelve.
You also need to see what happened, or you are guessing about your own app. How many opened it, how many came back the next day, and where they stopped. That is mobile app analytics, and at this size it means one event on the two screens that matter, not a dashboard. Ten installs plus one honest answer about why someone stopped is worth more than any chart you can build this month.
Try it
How many names is twelve testers
Your own guesses, not benchmarks. A one-to-one ask has no published average, so put in what you expect from your list.
5 testers from 40 asks
That is short of the 12 testers Google Play wants opted in for 14 days. At these rates you need about 96 asks, and the fix is a longer list rather than a better message.
On Android the first twelve users are a rule, not a milestone
Google wrote an answer to this question into its own requirements. A personal developer account created after November 13, 2023 has to run a closed test with a minimum of 12 testers who have been opted in continuously for at least 14 days before it can apply for production access. Until that is done, Production and Pre-registration stay switched off in Play Console.
Read what that means in order. You cannot be listed publicly until twelve people you found yourself have had the app for two weeks. Open testing, where anyone can join your test from Google Play, only becomes available after you gain production access. The route where strangers find you is closed at exactly the moment you want it open.
Google's own guidance for finding those twelve matches this page: personal and professional networks first, then communities where your future users already are, then your own followers. Internal testing has no entry requirement and builds are normally available to testers within seconds, so use it for the people who will tell you the truth before the fourteen day clock starts. You also have to summarise the testing feedback when you apply, so keep what they tell you.
Store search does matter later. A listing that converts, and the keyword work behind it, is app store optimization, and it starts paying once there are ratings and a screenshot worth looking at. It is not where the first users come from on either store, however much of the advice online is written as though it were.
Google Play Console Help, app testing requirements for new personal developer accounts
The install link decides more than the pitch does
Every ask ends with a link, and the steps behind that link cost you more people than your wording does. On iOS a beta goes out through TestFlight. Apple lets you invite up to 10,000 external testers, either by email or through a public link that needs no contact details from anyone, and you can set criteria such as device type and OS version for who joins by link. Your first build has to be approved by App Review for TestFlight before external testers can have it, and builds are sent for review automatically once you add them to a group.
Now count the steps for the friend you asked. Install TestFlight, accept the invite, install your build, open it, then find the thing you wanted an opinion on. Every step loses people who meant well. So shrink the ask. One screen, one question, and a time you will call them beats a request to try the app and tell you what you think.
Once you are public the link becomes a store link and the friction moves rather than disappearing. App launch users arrive the same way, one conversation at a time, with fewer screens in between. The store page is now doing part of the asking for you, which is the first point at which a listing is worth real effort.
If forty asks produce nothing, stop rewriting the message. The problem is usually what you built it for rather than the pitch, and the fix is a different app for the same people, not a louder launch.
Apple, TestFlight: inviting external testers and using public links
Where the first hundred can come from, and what each one costs
| How you reach them | Who it actually reaches | What it costs you | Works with no audience | Tells you why they left |
|---|---|---|---|---|
| Messaging people you can name | friends, colleagues, classmates | one message each, plus a follow-up | Yes | Yes |
| A community you take part in | people who have the problem | weeks of reading before you post | Yes | Yes |
| Posting to your own following | the audience you already had | minutes, if you have one | No | only if they reply |
| App store search | people typing a keyword | a listing, ratings and patience | No | No |
| Paid install ads | anyone you can target | money, before you know what a user is worth | Yes | No |
Getting it into ten hands this week
The shortest path to the first users is a build somebody can install today, not a store launch in three weeks. On iOS that is a TestFlight upload. On Android it is an internal testing release, which needs no minimum tester count. Both work while the app is still private, and both are where usable feedback comes from.
Newly is an AI app builder. You describe the app in plain English and it writes a real React Native and TypeScript project you own, runs it on cloud iOS and Android simulators while it builds, then uploads to App Store Connect and TestFlight or publishes to Google Play internal testing. It does not press submit for App Review on your behalf; you do that yourself. It is $25 a month and there is no free plan.
One detail matters for first users. A new app starts local-first, keeping its data on the phone with no accounts and no server, so there is nothing to sign up for, and that removes the step where most of an ask leaks away. Newly Backend gets added later, when the app needs accounts, shared data, syncing or payments. The code comes out as a ZIP from project Settings or through the two-way GitHub sync, so a tester link is never blocked on the tool.
Questions people ask about finding the first users
By asking named people one at a time, and by taking part in a place where people with the problem already talk. Write down everyone you can name who might want it, message them individually, and ask for one specific thing instead of general feedback. With no audience there is no faster route that is honest. Reach costs either money or years, and at this point you have neither to spend.
Write the list before you write the post
Put down every person you can name who has this problem, cut what you are asking them to do to one screen and one question, and get a build they can install while the conversation is still warm.
Start building