How to get your first 100 users, one person at a time

Why a hundred, and why by hand

A hundred is small enough to reach without marketing and big enough to show patterns. With ten users, every opinion is loud. With a hundred, you can see which kind of person keeps coming back and which kind leaves after a day, and that tells you who to find more of.

Doing it by hand is the point, not a phase to rush through. Paul Graham called it doing things that don’t scale: recruiting users one by one and giving each of them more attention than any company ever could later. Every conversation in this stage is also research, and you will not get it as cheaply again.

The ladder at a glance

The ranges overlap in practice. Keep messaging people while you post in communities, and keep both going through the launch.

Users 1 to 20: people you can name

Open a spreadsheet and write down fifty people who have the problem your product solves. Real names, not “freelancers” or “small teams”. Start with people you know, then add:

• Former colleagues and clients who did the job your product helps with.
• People you spoke to while validating the idea. They already told you about the problem.
• People who complained about a competitor in a review, a forum post or on social media.
• Friends of friends: ask five people “who do you know who struggles with this?”

Then message each one personally, a few a day. Mention why you thought of them, say what the product does in one line, and ask them to try it, not to “check it out”. Offer to set it up for them on a short call. Many will say no or not reply. That is normal. You only need about one in three.

If you have not spoken to customers yet, start there instead. Our guide to validating an idea in a week shows how to find ten people and what to ask them.

Users 20 to 50: go where they already talk

Your next users are already gathered somewhere, asking each other the question your product answers. Find two or three of those places: a subreddit for their trade, a Slack or Discord group, a forum, a LinkedIn or X circle, an indie maker community. Read them for a week before posting anything, so you know the rules and the tone.

Then be useful in public. Three things work well:

1. Answer questions your product helps with, fully, in the thread. Mention the product only when it is the honest answer, and say you made it.
2. Share what you learned building it: a short write-up, a number, a mistake. People engage with stories and ignore announcements.
3. Ask for feedback on something specific (the onboarding, the pricing page), not “thoughts?”. A specific ask gets specific replies, and some of those people sign up.

Most communities limit self-promotion, and some ban it outright. Read the rules, use the promotion threads where they exist, and never post the same message in ten places on the same day. One helpful reply in the right thread beats twenty links anywhere.

Users 50 to 100: launch, then launch again

With fifty people using the product, you have something to show: quotes, a clearer tagline, maybe a few numbers. Now a launch works, because the people who see it find a product that has already been tested, and your early users can vouch for it.

Do not bet everything on one day. Pick three to five places and spread them over a week or two, so each launch gets your full attention:

• Launch platforms and directories where founders and early adopters browse new products, LarkLaunch among them.
• Show HN on Hacker News, if your product is something people can try right away without a sales call or a long sign-up.
• Newsletters that feature new tools in your niche. Many take submissions for free.
• Your own list: every person from the first two steps, with a short note asking them to share it.

Our launch directories list shows the domain rating, traffic and price of each site, and the launch planner turns your date into a day-by-day schedule. For the details, read how to plan a launch week and how to write a launch listing people actually click.

Welcome every user by hand

Getting a sign-up is half the job. Most new users leave within the first session, before they reach the moment the product is useful. At this size you can prevent that personally.

• Write to each new user within a day, from your own address, as yourself. Ask one question: what made them sign up?
• Offer a short setup call to anyone who has not reached the first useful result. Watch where they get stuck; that is your next fix.
• Fix what they report fast, and tell them when you have. A user who reports a bug and sees it fixed the next day often tells others about it.

The replies to that one question are worth more than any analytics dashboard at this stage. They tell you, in your users’ words, what the product is for. Put those words on your landing page.

Know what worked

Before you share a link anywhere, tag it, so each sign-up tells you where it came from. Our UTM builder creates one tagged link per channel at once. After the hundred, compare channels on two numbers: how many people signed up, and how many were still active a week later. The second one matters more.

Once around forty people have used the product properly, send them the Sean Ellis survey, which asks how they would feel if they could no longer use the product. The usual benchmark is that 40% or more answering “very disappointed” is a strong early signal, and our product-market fit survey calculator scores the answers with the range the real number could be in. Below that, read the answers from the people who said “very” and build more of what they love.

Messages you can copy

Rewrite these in your own voice. Short and personal beats polished.

To someone on your list

Hi [name], you mentioned [problem] when we [worked together / spoke last month], so I thought of you.

I've built [product], which [one-line result]. It's early, and I'm looking for ten people to use it for real and tell me what's missing.

Would you try it this week? Happy to set it up with you on a 15-minute call.

A community post

I spent [time] trying to fix [problem] for [type of person]. Here's what I learned:

1. [Something specific and useful]
2. [A number or a mistake]
3. [What finally worked]

I ended up building a small tool for it ([product]). Happy to share it if it's useful, but mostly curious how you all handle this today.

The welcome email

Hi [name], I'm [your name], I made [product]. Thanks for signing up.

Quick question: what made you try it? Just hit reply, I read every answer.

If you'd like a hand getting set up, here's my calendar: [link].

Mistakes that stall at 30

• Waiting for the product to be ready. It never feels ready. If it solves the core problem badly, that is enough for the first twenty people, especially if you help them by hand.
• Broadcasting instead of messaging. One post to everyone gets less than ten messages to the right people.
• Counting sign-ups, not users. A hundred sign-ups and five active users is a product problem, not a reach problem.
• Spreading across ten channels. Pick two or three and do them properly. You will know within a few weeks which one is worth more.
• Launching too early. A launch with no users, quotes or clear tagline uses up a one-time chance. Get to fifty first.

The checklist

Questions founders ask