How to Get Your First 100 Users for an AI App

Wondering how to get the first 100 users for your AI app? The short answer: you do not buy them, you go get them by hand. The first 100 come from unscalable work: showing up where your niche already gathers, reaching out to people one at a time, and shipping something that looks credible enough that they are willing to try it. A free, polished design from VP0, the free iOS design library for AI builders, lets you ship a believable product fast, so your scarce hours go into users instead of pixels.

Who this is for

This is for solo founders and AI-assisted builders who have a working app and zero distribution. You do not have a budget for ads, you do not have an audience yet, and you need the first real humans using the thing this week.

Do things that do not scale

The single most useful idea here is Paul Graham’s Do Things that Don’t Scale: recruit your first users manually. That means direct messages, replies in the exact subreddit or Discord where your problem lives, a personal note to ten people who have complained about this problem in public. It feels slow because it is, and that is the point: each of the first 100 users teaches you something an analytics dashboard never will. The goal at this stage is learning, not volume.

Where the first 100 actually come from

Channel	Effort	Why it works for the first 100
Niche communities	Medium	People with the exact problem are already gathered
Direct outreach	High	A personal ask converts far better than a broadcast
A public build log	Medium	Sharing progress earns trust before launch
Launch directories	Low	A spike of curious, tolerant early adopters
Friends of users	Low	Happy users refer people just like them

Notice what is missing: paid ads. Andrew Chen’s widely cited data shows the average app loses about 77% of its daily active users within three days, so spending money to acquire users before you understand retention just fills a leaky bucket faster. Earn the first 100 by hand, watch who stays, fix what makes them leave, and only then consider paid channels.

Make the app worth sharing

Outreach fails when the product looks unfinished. Early users decide whether you are credible in seconds, and a rough, generic interface reads as untrustworthy no matter how good the underlying AI is. This is the practical reason design matters for growth: a clean, native-feeling app removes the biggest objection before you even make your pitch. Pick a design from VP0, copy its link, and have your AI builder rebuild it:

Rebuild this VP0 onboarding and home design in SwiftUI: [paste VP0 link]. Make the first run obvious, show the core value within one screen, and keep the visual polish consistent so the app looks trustworthy to a first-time user.

Then make sure you can learn from those users. Set up App Store Connect analytics before launch, and browse the practical playbooks in the Y Combinator startup library for getting traction. The polish-and-instrument loop pairs well with an AI prompt testing library and getting your store listing right with App Store screenshot dimensions for 2026. If your app is a thin layer over a model, keep the keys safe as shown in the OpenAI API wrapper app template. And once users arrive, keep them engaged with patterns like a leaderboard podium animation.

Common mistakes

The first mistake is launching to no one: posting once to an empty audience and calling it a launch. The second is buying ads before you know who retains. The third is shipping with no analytics, so the first 100 users teach you nothing. The fourth is a rough interface that makes outreach feel like an apology. The fifth is treating 100 as a vanity number instead of 100 conversations that tell you what to build next.

Key takeaways

• The first 100 users come from unscalable, human outreach, not ads.
• Go where your exact niche already gathers and message people one by one.
• Ship a credible-looking app first; rough design kills early trust.
• Use a free VP0 design to look real fast and spend your time on users.
• Instrument before launch so every early user teaches you something.

Frequently asked questions

How do I get my first 100 users for an AI app? Go to them directly: post in the niche communities where your problem lives, message relevant people one at a time, and ship an app polished enough to share. The first 100 are won by hand.

Should I run ads to get my first users? Almost never at the start. Without retention data, ads buy noise. Earn the first 100 manually, learn what makes them stay, then test paid channels.

Can VP0 help me launch an AI app for free? Yes, indirectly. VP0 is a free iOS design library, so you can ship a credible-looking app at no cost, which makes your outreach far easier.

How important is the app’s design when getting early users? Very. Early users judge credibility in seconds, so starting from a strong free design removes the rough-edges objection before you pitch.