Your first 100 users without a launch spike

A practical way to find your first 100 users by finding the people who already have the problem, before and after launch.

A launch spike is the hard way to get your first users: you compete with everyone else’s announcement for a day. There is a slower, steadier way that usually works better for small products. Find the people who already have the problem, and talk to them one at a time.

This guide is about that. It assumes you have something that works, even roughly.

The one idea: look for the problem, not for an audience

An audience is people who follow you. A user is someone who has the problem and would use your product to solve it. For a new product there are far more of the second group than the first, and they are easier to reach, because they are already complaining, asking and searching.

So the job is not “get attention”. It is “find a person who has this problem today, and show them”.

Step 1: Describe your user in one sentence

“Freelancers who invoice in more than one currency.” “Solo developers who run their own monitoring.” “Small clinics that still book on paper.”

If you cannot say it in one sentence, you will not be able to find them. Narrow it until it feels almost too small. You can widen later.

Step 2: Find where those people already talk

For almost every kind of user there are three places:

Communities and forums where they ask each other for help: subreddits, Slack and Discord groups, niche forums, Stack Exchange sites. Read each community’s rules on self-promotion before you post; many restrict it.
Comment sections and threads under tools they already use, where people list what is missing.
Search results for the problem phrased the way they would phrase it.

Spend a week reading, not posting. Write down the exact words people use for the problem. Those words are your headline.

Step 3: Help first

When you find someone with the problem, answer the question they asked. Only mention your product if it genuinely fits, and say that you built it. People tolerate a maker who is useful and honest about being a maker. They do not tolerate a pitch dropped into a thread.

A good test: if your product did not exist, would the reply still be worth writing? If yes, write it.

Step 4: Talk to the first ten by hand

Your first ten users are not a growth channel, they are research. For each one:

Ask what they were trying to do when they found you.
Watch them use it, if they allow it.
Write down where they got stuck.

Do this before you automate anything. Ten conversations will tell you more than a thousand visits.

Step 5: Make yourself easy to find

While you do that, quietly build the pull side so you are not starting from zero each week:

Put a plain, specific sentence on your homepage (“Invoicing for freelancers who work in several currencies”).
Add the product to places that keep listing you, not only places that feature you for a day. See where to list your SaaS.
Write down the answers to the questions you keep being asked, and publish them as short pages. They become the thing someone finds later.

Step 6: Count what matters

A hundred users is a modest number, so you can track it by hand. For each person, note:

Where they came from.
Whether they got to the moment where the product worked for them.
Whether they came back.

If most of your users come from one place, do more of that. If none of them come back, the problem is the product, not the marketing, and that is useful to know early.

What not to do

Buy ads before you know who buys. You will pay to learn what a conversation would have told you for free.
Chase vanity numbers. Followers and upvotes are not users.
Mass-message strangers. It costs trust and rarely works.
Wait for a big launch. Start talking to people now. A launch is more useful once you have real users to point to.

Keep reading

Updated 4 October 2026.