The first 100 users are not a growth milestone.

They are a learning instrument.

Most founders misunderstand early users. They want the first 100 to prove the business is working. That is too much pressure and not enough precision.

The first 100 should help you answer:

Who cares?
Why do they care?
What did they expect?
Where did they come from?
What did they do?
What confused them?
What would make them stay?
What would make them pay?

Growth comes later.

First, learn.

The first-users principle

Your first users should be close enough to teach you.

That does not mean friends who are polite. It means users whose behavior you can observe, question and understand.

Paul Graham’s “Do Things that Don’t Scale” is often interpreted as a growth hack. It is better understood as a learning doctrine. Early manual work gives founders access to details they would miss if they tried to automate too soon.

The first 100 users are where you earn those details.

Do not start with scale

Avoid the beginner fantasy:

Launch publicly.
Go viral.
Acquire users.
Scale.

The better sequence:

Identify a specific user.
Reach them directly.
Watch what they do.
Fix what breaks.
Repeat.

Scale hides truth. Direct contact reveals it.

Source 1: Direct outreach

This is the least glamorous and most useful source.

Make a list of people who have the problem.

Where to find them:

  • LinkedIn search;
  • niche communities;
  • Twitter/X;
  • Reddit;
  • Slack groups;
  • Discords;
  • newsletter comments;
  • app reviews;
  • competitor communities;
  • personal network;
  • conference speaker lists;
  • job postings.

Do not send generic spam.

Send a specific note:

I’m building a small tool for [specific audience] who struggle with [specific problem]. 
I noticed [specific context]. 
Would you be open to trying an early version or telling me how you solve this today?

The goal is not to close. The goal is to start conversations.

Source 2: Communities

Communities can work if you enter with respect.

Do not arrive with a launch link.

Arrive with a useful artifact:

  • checklist;
  • teardown;
  • template;
  • benchmark;
  • guide;
  • comparison;
  • calculator;
  • honest build log;
  • question with context.

Bad community post:

I built this app. Check it out.

Better:

I analyzed 50 landing pages for side-business tools and found 7 patterns. Here is the breakdown. I also built a free checklist if useful.

Communities reward contribution before extraction.

Source 3: Content

Content can acquire first users if it is specific enough.

Do not write:

How to start a business.

Write:

How to validate a side business while working full-time.

Do not write:

Best AI tools.

Write:

The AI workflow stack for solo consultants doing client research.

Content works when it carries intent.

At the start, optimize for quality of reader, not quantity of traffic.

Source 4: Paid tests

Paid tests can accelerate learning.

Small budgets can test:

  • search intent;
  • landing page promise;
  • audience resonance;
  • pricing;
  • creative angle;
  • app-store page quality;
  • newsletter signup cost.

But paid traffic should not replace conversations.

If you buy 500 visitors and learn nothing about why they acted, you only bought numbers.

Use paid tests to create conversations and actions.

Source 5: Founder-led sales

Some businesses require founder selling.

This is not a failure.

Founder-led sales teaches:

  • buyer objections;
  • urgency;
  • budget owner;
  • decision process;
  • language;
  • alternatives;
  • must-have features;
  • trust gaps.

Even if the final product is self-serve, early sales calls can reveal what the landing page cannot.

The first 100 map

Break the first 100 into four stages.

Users 1–10: learning

Goal:

Watch behavior closely.

Actions:

  • onboard manually;
  • ask questions;
  • observe confusion;
  • collect objections;
  • fix obvious issues.

Do not scale.

Users 11–30: pattern recognition

Goal:

Find repeated signals.

Actions:

  • segment users;
  • track activation;
  • identify best-fit users;
  • test pricing language;
  • improve onboarding.

Look for:

  • who gets value fastest;
  • who complains most usefully;
  • who would be disappointed if it disappeared.

Users 31–70: channel testing

Goal:

Find a repeatable acquisition source.

Actions:

  • test 2–3 channels;
  • compare conversion quality;
  • track retention by source;
  • identify channel-user fit.

Do not only track acquisition cost. Track quality.

Users 71–100: decision

Goal:

Decide what to do next.

Actions:

  • write evidence memo;
  • choose core audience;
  • kill weak features;
  • improve one channel;
  • decide whether to scale, pivot or narrow.

What to track

For the first 100 users, track:

Source
User type
Problem
Action taken
Activation
Feedback
Payment intent
Retention/repeat usage
Objection
Next step

A spreadsheet is enough.

Do not overbuild analytics before you know what matters.

The 100-user interview questions

Ask:

What were you trying to do when you found this?
How do you solve it today?
What almost stopped you from trying it?
What did you expect to happen?
Where did the product feel unclear?
What would make this worth paying for?
Who else has this problem?
What would you search for if you needed this?
Would you be disappointed if this disappeared?

The last question is uncomfortable. Good.

The first-users mistake

The biggest mistake is chasing users who cannot teach you.

Bad early users:

  • friends being nice;
  • giveaway seekers;
  • random traffic;
  • people outside your target;
  • users who came for the wrong promise;
  • people who like the idea but do not have the problem.

Good early users:

  • people with the pain now;
  • people using alternatives;
  • people willing to give feedback;
  • people willing to pay;
  • people who repeat the behavior;
  • people who refer others with the same problem.

The founder lesson

The first 100 users are not there to validate your ego.

They are there to sharpen the business.

Treat them like a temporary advisory board created by the market.

Talk to them. Watch them. Serve them manually. Learn their language. Track what they do. Pay attention to who returns.

The first 100 users should not make you famous.

They should make you less wrong.

References