How to find your first 100 customers

Your product works. The signup page is live. And nobody is signing up. This is the part nobody warns you about: building was the easy half. The first 100 customers are the hardest you will ever get, because you have no reputation, no reviews, no referrals, and no momentum. You earn them one at a time, by hand.

The good news is that getting 100 people to care is a tractable problem. It is mostly effort, not genius. The founders who succeed here are not the ones with the cleverest growth hack. They are the ones willing to do unglamorous, manual work for months while the channels that compound slowly start to catch.

Kill "build it and they will come" first

This belief is responsible for more dead startups than bad code. The assumption is that a good product creates its own demand, that quality is self-advertising. It is not. The world is full of excellent products nobody has heard of, and mediocre ones with great distribution that print money.

Distribution is a separate job from building, and it does not start after launch. It starts before. If you have not already done this, validate the idea first by talking to the people you expect to pay you. Those conversations are not just research. They are your first leads. The person who told you your idea was interesting is far more likely to try it than a stranger you reach later. Treat every pre-launch conversation as a future customer, not a focus group.

Go where your users already gather

You do not need to invent an audience. For almost any product, the people who have your problem are already congregating somewhere and talking about it. Your job is to find those rooms and show up as a useful member, not a billboard.

That means the relevant subreddits, the niche Discords, the Slack communities, the old-school forums that still run the industry, the Facebook groups, the Twitter or LinkedIn circles where your users complain about exactly the thing you fixed. Find them, join them, and read before you post. Spend a week answering questions and being helpful with no link in sight. When someone describes the pain your product solves, then you mention what you built, plainly, as one option.

This is slow and it does not scale, which is precisely why it works. Communities have finely tuned spam detectors and almost no tolerance for drive-by promotion. The founder who has been genuinely helpful for two weeks gets a pass. The one who shows up to drop a link gets banned. The mapping work pays off too: a clear list of every community where your users live is a reusable asset. This is the sort of landscape you can sketch quickly with Kalit Search, which researches a market and surfaces where a given audience clusters, so you start with a map instead of a search bar.

Do things that don't scale

Paul Graham's advice has become a cliche, which is a shame, because most founders nod at it and then do the scalable thing anyway. Doing things that don't scale means manual, personal, one-to-one work that you could never sustain at 10,000 customers, and that is exactly why it is your edge at 10.

Reach out to people individually. Send a real DM that references something specific about them, not a copy-paste blast. Write cold emails that read like a human wrote them to one recipient, because one human did. Onboard your first users over a call. Fix their problem live. Send a handwritten thank-you if you have to. The goal is not efficiency. The goal is a small number of people who feel personally looked after and tell others. Cold outreach done well converts; done lazily it is just noise you are adding to an already noisy inbox.

Launch on the obvious places, with realistic expectations

Product Hunt, Hacker News, Indie Hackers, the big relevant subreddits. These are worth doing, but recalibrate what a launch actually delivers. A launch is a spike, not a foundation. You get a day or two of traffic, some signups, occasionally a useful piece of feedback or press, and then it goes quiet. People who expect a launch to make their business are reliably disappointed.

Treat each launch as one event in a sequence. Prepare properly: a clear one-line description, a demo people can grasp in ten seconds, and answers ready for the comments. Show up and engage all day. Then do it again somewhere else next month. The compounding does not come from one big bang. It comes from being present in five or six places over a quarter, with each appearance adding a handful of users and a little credibility.

Capture intent with a waitlist and a landing page

Most of the traffic from any channel is not ready to commit the moment they arrive. If your only options are "sign up now" or "leave forever," you lose the maybes, and the maybes are most of them. A landing page that captures email gives you a way to keep the conversation going.

A waitlist also tells you something honest before you have built everything: whether anyone actually wants this. The signup rate against your traffic is a real signal, far better than compliments. Building one is its own small craft, and the difference between a page that collects emails and one that collects nothing is mostly clarity and proof. Here is how to build a waitlist page that actually converts. Capture intent, then nurture it with occasional, genuinely useful updates rather than silence followed by a desperate launch blast.

Turn early users into referrals

Your first customers are worth more than the revenue they bring, because each one can introduce the next. A happy user with a warm network beats any ad. But referrals rarely happen by accident; you have to ask, and you have to make it easy.

Ask directly: "Do you know one other person who has this problem?" Most people are happy to help if you make the request specific and low-effort. Give them something to forward, a sentence they can paste, a reason their colleague would care. You can offer a small incentive, but for the first 100, the stronger lever is simply that you have delighted someone enough that they want to look good by recommending you. Earn that, then ask for it.

Treat content and SEO as the slow compounding bet

Everything above is push: you going out and pulling people in by hand. Content and SEO are the pull channel, where people find you because they searched for the problem you solve. It is the channel founders most want and most underestimate, because it pays nothing for months and then, if you are consistent, quietly becomes your largest source of qualified signups.

Write the things your users are searching for: how to solve the problem, comparisons, the questions they ask before they know your product exists. It will feel pointless at first. The articles get no traffic, nothing happens, and you will be tempted to quit right before it works. Plant it early so it has time to grow while you do the manual work that gets you to 100 today.

To recap, the path to your first 100 customers:

1. Abandon "build it and they will come" and treat distribution as its own job, starting pre-launch.
2. Find the communities where your users already gather, and be useful there before you sell.
3. Do things that don't scale: personal outreach, real cold email, hands-on onboarding.
4. Launch on the obvious platforms, but expect spikes, not foundations, and launch repeatedly.
5. Capture intent with a landing page and waitlist instead of losing the maybes.
6. Ask delighted users for one specific referral and make it effortless.
7. Start publishing content now so the slow compounding channel is working by the time you need it.

None of this is clever. It is just work, done earlier and more personally than your competitors are willing to do it. The first 100 customers reward stubbornness, not brilliance.