How to find the right users for an app you built with AI
You built the app. It works. You posted it once, a few friends said "nice", and now the analytics show a flat line.
This is the most common place for a side project to stall, and it's usually not because the app is bad. It's because it hasn't met the people it's for yet. Finding them is a skill, and like building, it gets much easier once you know the steps.
This guide is about finding the right users: the first 10 to 100 people who have the problem your app solves and will actually use it. Not traffic, not likes. People who come back.
Start with the problem, not the app
Most builders describe their app by what it is: "an AI-powered scheduling tool with a clean UI." Nobody searches for that. People search for what's going wrong: "clients keep missing appointments," "how do I take a deposit when someone books."
Before you look for anyone, write one sentence:
[Who] who [has this problem], and right now they [do this clumsy workaround].
For example:
- Yoga teachers who lose track of who booked which class, and right now they keep a list in their notes app.
- Indie developers whose coding agent ignores their design, and right now they paste screenshots and explain it in a paragraph.
- YouTubers who need short clips from long videos, and right now they cut them by hand in a desktop editor.
If you can't fill in that sentence, you're not ready to look for users yet. You'd be looking for "anyone," and anyone is no one.
The workaround part matters most. A painful workaround means the problem is real. If people aren't doing anything about it today, they probably won't switch to your app either.
Find where they already talk about the problem
Your users are already describing their problem somewhere, in their own words. Your job is to find those places.
- Reddit and niche forums. Search the problem, not your category. "missed appointments" in a subreddit for salon owners. "invoice" in a freelancer community. Read the threads where people ask for help.
- X and Bluesky. Search the complaint: "anyone know a tool that," "I wish there was an app," "so tired of." Recent posts are people who have the problem right now.
- Reviews of the big tools. The 2- and 3-star reviews of the incumbent in your space are a list of what people wish it did. If your app does that one thing well, those reviewers are your users.
- Search autocomplete. Type your problem into Google and note what it suggests. Those are the exact phrases people use. So are "alternative to [big tool]" searches.
- Your own life. If you built this for yourself, the people around you with the same job (coworkers, friends in the same trade) are the fastest first users you'll get.
Write down the exact words people use. You'll need them for the next two steps.
Talk to them one by one
The first users almost never come from a big launch. They come from one-to-one conversations. This feels slow and unscalable. It is, and that's fine. You only need ten.
When you find someone who described the problem:
- Reply about their problem, not your app. "I had the same issue with no-shows. I ended up building something that takes a deposit when people book. Happy to set it up for you if it helps."
- Offer to do the setup for them. Getting someone to try a new tool is hard. Getting them to accept a favor is easy. Setting it up together also shows you exactly where they get confused.
- Ask, don't pitch. "What do you use for this today? What's annoying about it?" The answers are worth more than the signup.
Some people will ignore you. A few will say yes. Those few are the most valuable users you'll ever have: they'll tell you what's wrong, and they'll forgive you while you fix it.
Make trying it take ten seconds
Every step between "this looks interesting" and "I'm using it" loses people. With your first users, cut every step you can.
- Let people try before signing up. A demo, a sample project, a guest mode. Ask for an email once they've seen it work, not before.
- Put their words in your headline. If people say "clients keep missing appointments," your headline says that, not "AI-powered scheduling."
- Show it working. A 20-second video or GIF of the core action beats any amount of copy.
- Have a real URL that loads fast on a phone. A lot of people will first see your app on their phone, from a link someone sent them.
Be findable by the problem, not just your name
One-to-one outreach gets your first ten users. To keep getting more without doing it forever, you need to show up where people go when they have the problem.
- Write one page per problem. A page titled with the exact phrase people search ("How to stop clients missing appointments") that explains the problem honestly and shows your app as the fix. One good page can bring in people for years.
- List it where people browse by problem. Directories where people search by what they need, not just by what's new. That's what we built JustBuildApps for: people type their problem in their own words and get the app that does it. Your listing doesn't expire after a day.
- Make it easy for AI assistants to recommend you. More people now ask ChatGPT, Claude or Cursor "what's a tool that does X" before they search Google. Have a clear one-sentence description of what your app does and who it's for on your homepage. Add an
llms.txt. Get listed in catalogs that agents can search, like ours through our MCP server.
Launch, but treat it as one channel
Launching on Product Hunt, Show HN or Reddit is worth doing. Just know what it is: a spike of attention for a day or two, mostly from other builders.
That last part is easy to miss. Launch communities are full of makers, not your customers. A salon owner isn't browsing Product Hunt. Other builders will give you useful feedback and encouragement, but if your users are yoga teachers or freelancers, most of them are somewhere else.
So launch, collect the feedback, thank everyone, and then go back to where your actual users are.
Learn from the first ten
Once a few people are using it, they're your best source of what to build next and where to find more people like them.
Ask every early user three questions:
- "What were you trying to do when you found this?" This tells you the real job, which is often different from what you built it for.
- "What almost stopped you from trying it?" This tells you what to fix on your homepage.
- "Where would someone like you hear about a tool like this?" This tells you where to find the next hundred.
And if you can, watch one person use it without helping them. It's painful and it's the fastest way to find what's confusing.
Measure people who come back, not signups
Signups feel good and tell you very little. The number that matters early is: did people use it again a week later?
If a handful of people come back on their own, you have something. Keep going and find more people like them. If nobody comes back, more traffic won't fix it. Talk to the people who left and find out why.
Common mistakes
- Building for everyone. "Anyone who needs to be productive" isn't a user. Pick one group and win them first.
- Waiting until it's perfect. Your first users will change what you build. Show it to them while it's rough.
- Only posting where other builders hang out. Fun, encouraging, and usually not where your users are.
- Pitching instead of listening. People share their problems freely. They tune out pitches.
- Giving up after one launch. One quiet launch day says almost nothing about whether people want your app.
Where JustBuildApps fits
We built JustBuildApps because finding users is the part that kills most small apps, and we think it shouldn't be.
On JustBuildApps, people find apps by describing their problem. They can try your app right on the page in one tap, and like, save, share or follow you. Build requests show you problems people have that no app solves well yet. Your listing keeps working for you long after launch day.
It's free. If you've built something, post it. If it solves a problem someone has, they'll find it.
Questions people ask
How do I get my first users for a new app?
Write down exactly who has the problem your app solves, find where they already talk about that problem (Reddit, X, niche forums, reviews of bigger tools), and reach out one by one. Offer to set the app up for them. Your first ten users almost always come from direct conversations, not a launch.
Where can I find users for my app for free?
Communities where your users discuss their problem, replies to people asking for a tool on X or Reddit, problem-focused pages on your own site that rank in search, and app directories where people browse by problem, like JustBuildApps. All of these are free; they cost time instead of money.
Should I launch on Product Hunt?
Yes, it's worth doing, but treat it as one channel. A launch gives you a day or two of attention, mostly from other builders. If your users aren't builders, most of them won't be there, so go back to where your actual users spend time afterward.
How many users do I need before I know my app is working?
Fewer than you think. If ten people who aren't your friends use it and a few come back the next week on their own, you have something worth growing. If nobody comes back, more traffic won't help; talk to the people who left.
How do I get AI assistants like ChatGPT or Claude to recommend my app?
Say clearly on your homepage what the app does and who it's for, in one sentence. Add an llms.txt file. Get listed in catalogs that AI agents can search, such as JustBuildApps through its MCP server. Assistants recommend what they can find and understand.
Built something?
Put it where people with that problem are looking. It's free.
~JustBuildApps