You Shipped Your AI-Built App. Now What? How to Get Your First 100 Users

AI coding tools removed the hardest-looking part of starting a product. They did not remove the part that decides whether it survives: getting the right people to try it. Founder communities are full of the same post in different words: "I built it, it works, and I have zero users."

The good news is that the first 100 users rarely come from clever growth hacks. They come from a handful of boring, repeatable actions done every day for a few weeks.

Why a working product still gets no users

• The audience is "everyone". If you cannot name where your users spend time online, you cannot reach them.
• The promise is a feature list. People look for an outcome ("send invoices in 30 seconds"), not a tool ("an AI-powered invoicing platform").
• One launch post, then silence. A single post gets a spike and then nothing. Distribution is a routine, not an event.
• No way to come back. Visitors who are not ready today leave forever because there is no email capture, no content and no follow-up.

Step 1: Narrow the audience until you can list 50 people

Write down one sentence: "This is for [role] who [situation] and currently [workaround]." Then try to list 50 real people or companies that fit. If you cannot, the audience is too vague. Narrow it until you can. Those 50 are your first outreach list.

Step 2: Direct outreach, ten messages a day

The first users almost always come from one-to-one messages, not broadcasts. A message that works:

• mentions something specific about the person or their work;
• names the problem in their words, not your product's features;
• asks for feedback or a 15-minute call, not a purchase;
• is short enough to read on a phone.

Ten personal messages a day for three weeks is about 200 conversations started. Even a modest reply rate gives you dozens of real users and, more importantly, the words they use to describe their problem.

Step 3: Show up in communities as a helper, not an advertiser

Reddit, Indie Hackers, niche Slack and Discord servers, Telegram chats and LinkedIn groups are where your audience already asks for help. The pattern that works:

1. Read the rules. Many subreddits ban links or limit self-promotion to specific threads.
2. Answer questions thoroughly for a week or two before mentioning anything you built.
3. Share what you learned building it ("how I cut onboarding from 6 steps to 2"), with the product as context, not the headline.
4. Reply to every comment. Engagement keeps the post visible and earns trust.

Step 4: Use launch platforms as a multiplier

Product Hunt, Show HN on Hacker News, Indie Hackers and startup directories can bring a burst of visitors and early adopters. They work best when you already have a few happy users, a clear one-line promise, screenshots or a short demo video, and a free way to try the product immediately. Treat them as one week of your plan, not the whole plan.

Step 5: Plant pages that keep working while you sleep

Outreach does not scale; search does. Write a few pages that answer the exact questions your audience types into Google and AI assistants: "how to [task] without [pain]", "[competitor] alternative", "[task] template". Each one should solve the problem on the page and offer your product as the faster way. See how to get cited by ChatGPT, Perplexity and Google AI and why Google is not indexing your new site.

Step 6: Turn every user into proof

• Ask each early user what changed for them and quote it, with permission, on your site.
• Post short before-and-after examples and real numbers.
• Build in public: weekly updates on what you shipped and learned attract people who want to follow along.

Step 7: Measure the funnel, not the vanity

Track four numbers weekly: visitors, sign-ups, users who reached the key action, and users who came back. If visitors are high and sign-ups low, fix the promise. If sign-ups are high and activation low, fix onboarding. If activation is fine and nobody returns, fix the product.

A 30-day plan

• Week 1: narrow the audience, list 50 people, rewrite the headline as an outcome
• Week 1–4: ten personal messages a day, every reply logged
• Week 2: start answering questions in three communities
• Week 3: publish two search pages that solve real questions
• Week 4: launch on one platform with quotes from first users
• Every week: review the four funnel numbers and change one thing

Week five of the Ship Index Grow program is built around exactly this: each participant leaves with a distribution routine and real conversations, not just a deployed app. If you have not shipped yet, start with the six-week launch playbook.

Questions people ask

Why does my app have no users even though it works?

Usually because nobody knows it exists or the people who see it do not recognize their problem in it. A working product is the entry ticket; distribution and a clear promise to a specific audience are what bring users.

Where can I launch a new app for free?

Product Hunt, Hacker News (Show HN), Indie Hackers, relevant subreddits and niche communities, BetaList-style directories and your own LinkedIn or Telegram audience. Read each community's rules first; most ban pure self-promotion.

How long does it take to get the first 100 users?

With daily outreach and a clear audience, many small products reach their first 100 sign-ups within one to two months. Without a defined audience it can take far longer, no matter how good the product is.

Should I pay for ads to get first users?

Not at first. Early on you need conversations and feedback more than traffic. Ads make sense later, once you know which message converts and what a user is worth.