How to get your first 10 customers

The short answer

Get your first 10 customers by talking to them one at a time. Start with people you know, move to people already complaining about the problem in public, ask about their problem before you mention your product, and charge from the first conversation. Ads, SEO and virality come later, because at ten customers you still need to hear why people buy.

This is older and more proven than any growth hack. Paul Graham wrote in Do Things That Don't Scale ↗ in July 2013 that recruiting users one at a time is the most common unscalable thing founders have to do, and that nearly all startups have to do it. Airbnb's founders went door to door in New York. Stripe's would answer a yes with "Right then, give me your laptop" and set the person up on the spot. YC named that the Collison installation.

Lenny Rachitsky's July 2020 review ↗ of how fast-growing B2B companies found their first ten customers lands in the same place: personal network, going to where customers already are, and press, and he notes that press is rarely how you get started. That leaves two doors, and both are open to you today.

What founders actually did

Forget the case studies with a funding round in paragraph two. These come from founders answering the question in public, in their own words.

Look at what they share. No big launch, no ad budget. Each one is a person choosing another person and writing to them. Des Traynor, who co-founded a company most of you know, got up at 4am to present to nine people. And look at that cold-calling line. Two hundred noes is the price some founders pay, and they keep going because each no costs less than building another feature nobody asked for.

Find the people who already have the problem

Start with 30 names

Open your contacts, LinkedIn and old inboxes. List former colleagues, past clients and anyone who works where your user works. PostHog's James Hawkins ↗ began by mining LinkedIn for relevant contacts, including former coworkers. You do not need them to buy. You need them to buy or to name someone who might.

Then search for pain, not for profiles

People describe their problems in public every day. Search Reddit, X, Hacker News and niche forums for what a frustrated user would type, not what a marketer would type:

• "alternative to [competitor]"
• "[competitor] too expensive" or "[competitor] doesn't work"
• "how do you handle [task]" or "is there a tool for [task]"
• "I'm so tired of [task]"

Then pick your threads. The founder who called Reddit a "problem database" in February 2026 ↗ went for threads with 0 to 20 comments, accepted 20 to 50, and skipped anything over 100, because an answer under 150 replies is invisible. New threads with a frustrated person in them are where one good reply gets read.

The Next Web's 2015 piece ↗ put it plainly: "A competitor's unhappy customer is a hot lead for your company." Reply in public when it truly helps and the community allows it. Our community promotion guide shows you how to stay welcome while you do.

What to say, and what never to say

The question that kills first conversations is "would you use this?" People are polite, so they say yes and then vanish. Rob Fitzpatrick wrote The Mom Test ↗ about exactly this: learning whether your business is a good idea when everyone is lying to you. PostHog's advice ↗ follows it, asking for specifics instead of "Would you buy this?". Ask what they did last time the problem happened, what it cost them, and what they tried.

Notice how the second message offers to set it up with them. That is the Collison installation for 2026: share your screen, get them to their first result, and do not leave them alone with a signup form.

Charge from the first conversation

A free user tells you the product is fine. A paying user tells you it is worth something. One founder in a 2017 Hacker News thread ↗ reported that a company paid a deposit before the company even existed. Another spent 9 months in beta with about 50 users, switched on pricing, and got the first 10 customers from that group. Those nine months are the part you can skip.

• Say the price out loud, early. A quick no saves you weeks of follow-ups.
• Pick one plan and one number. Your first ten customers will tell you if it is wrong.
• Offer a founding price in exchange for a weekly call. It is honest and it makes you a better product.
• Track what you earn from the first sale with the MRR calculator. Watching churn at ten customers beats discovering it at a hundred.

The routine that gets you there

PostHog co-founder James Hawkins set himself 2 meetings a day, at about 2 hours daily ↗, with each conversation leading to 2 more contacts. The arithmetic in that article comes to 10 prospects a week and 40 a month. You may not get two meetings a day when you start, so here is a smaller version of it, our own suggestion for one person with a day job:

1. Five new messages a day, each one naming something specific about the person.
2. One public reply where you can genuinely help.
3. One conversation a day, booked or chased.
4. End every conversation with: "Who else deals with this?" A founder in the same 2017 thread ↗ recommended asking everyone you meet for a referral.
5. Write down the exact words people use for the problem. They become your homepage.

Make sure they can check you out

Here is what happens after your good cold message: the person searches for your product. If they find one lonely homepage, your message just became less believable. If they find your product listed in places they recognize, it looks like something that exists.

That is the reason to sort out listings now rather than at customer one hundred. It costs you a few hours or one small payment, and those pages stay up while you are busy having conversations.

Frequently asked questions