Everyone wants the growth-hacking answer to this. The viral post, the clever ad, the automated funnel that prints customers while you sleep. For your first ten customers, none of that works, and chasing it is why most new businesses stall before they've sold anything.

The honest version: the first ten customers are manual. You get them one conversation at a time, by hand, often by people who already know and trust you. That feels too slow and too unglamorous to be the real answer. It is the real answer. Here's how to actually do it.

Why the first ten are manual, not scalable

This isn't just a vibe. It's the most common advice given to early-stage founders for a reason. In his essay Do Things That Don't Scale, Paul Graham makes the case plainly: nearly all startups have to recruit their first users manually, and you can't sit back and wait for them to come to you.

The examples are now famous. Airbnb's founders went door to door in New York signing up hosts and helping them fix their listings. The Stripe founders, when someone agreed to try the product, would say "give me your laptop" and set it up on the spot rather than emailing a sign-up link. That's the energy. Embarrassingly hands-on, completely unscalable, and exactly what gets you off zero.

The unscalable stuff isn't a phase you tolerate before the real growth. It's the thing that teaches you what to build, who to sell to, and what they'll actually pay for.

Trying to automate before you've sold by hand is like designing a factory before you've made one thing someone wanted. You don't know the product well enough yet. The first ten customers are how you learn it.

Start with your warm network

Your fastest customers are people who already trust you. Not because they'll buy out of pity — that's a weak first sale anyway — but because they'll actually take your call, give you honest feedback, and tell you when something's off.

Make a list. Genuinely write it out: former colleagues, old clients, friends in adjacent industries, people who've asked what you're working on. Then reach out individually. Not a mass announcement to your whole feed — a specific, personal message to one person at a time.

  • Be specific about who you help. "I'm doing X for people like you" lands far better than "I started a business, check it out."
  • Ask, don't pitch. "Is this a problem for you, or do you know someone it's a problem for?" opens a door. A hard sell slams it.
  • Make it easy to say yes. Offer to just do the thing, simply, for a fair price, with no contract theater.

Some of these become customers. Others become the introductions that lead to customers, which matters just as much.

Go where your people already gather

Once the warm network is tapped, go to the places your ideal customer already hangs out. Niche communities beat broad broadcasting every time when you're this small. A focused Facebook group, a subreddit, an industry Slack or Discord, a local meetup, a forum for the exact problem you solve.

The rule that keeps you from getting banned and ignored: be useful before you're promotional. Answer questions. Share what you actually know. Be a real person in the space for a bit. When someone describes the exact problem you solve, that's when you speak up — helpfully, not with a billboard.

Ten genuinely engaged people in a community where your customer lives are worth more than ten thousand strangers who scrolled past an ad. Depth over reach, every time, at this stage.

Direct outreach, done like a human

Warm network and communities won't always cover ten. The rest often come from direct outreach — messaging people who fit your customer profile but don't know you yet. This is where most people either chicken out or do it so badly it backfires.

The difference between outreach that works and spam that gets you blocked:

  • Research before you send. One sentence proving you actually looked at who they are beats a hundred copy-paste blasts.
  • Lead with their problem, not your product. Open with something true about their situation, not a feature list.
  • Make a small, clear ask. A quick call, a free look at their thing, an honest question. Not "buy my package" in message one.
  • Send a handful with care, not hundreds on autopilot. At ten customers, a high-touch trickle outperforms a high-volume firehose. You're not running a campaign, you're starting conversations.

Make the offer simple and the yes easy

A confusing offer kills early sales faster than a bad product. When you're new, simplicity is your advantage. One clear thing, one clear price, one clear result. Don't present three tiers and an add-on menu to someone who's never bought from you. That's a decision they'll defer, and a deferred decision is a lost sale.

You can also lower the risk of the first yes without giving the work away. A small pilot, a money-back promise, a single deliverable instead of a long commitment. The goal is to make saying yes feel safe, because they have no track record with you yet. Don't undersell yourself permanently — early discounts are fine, free forever is a trap — but do make that first step low-stakes.

Ask for referrals, properly

Once you've made even two or three people genuinely happy, you're sitting on your best growth channel and probably not using it. People rarely refer you unsolicited, not because they don't like you, but because it doesn't occur to them. So ask.

  • Ask right after a win, when they're visibly pleased with what you delivered. That's the moment, not three months later.
  • Make it specific. "Do you know one other person who's dealing with X?" is answerable. "Know anyone who needs my services?" gets a polite shrug.
  • Make the intro effortless. Offer to write the message they can forward. Remove the work.

A happy customer with a warm introduction is the highest-converting lead you'll ever get. The first ten, done right, should be quietly generating the next ten.

The honest part

This is slower and more uncomfortable than the version where customers magically appear from a clever tactic. You'll send messages that get ignored. You'll have conversations that go nowhere. You'll feel like you're begging, especially the first few times. That feeling is normal and it fades fast once people start saying yes.

Here's the payoff that makes the grind worth it: every one of those manual conversations teaches you something a dashboard never could. Why people hesitate. What objection keeps coming up. Which words make them lean in. By customer ten, you don't just have ten customers — you understand your business in a way you couldn't have from any amount of planning. That's the asset. The customers are almost a bonus.

Scaling comes later, and it comes easier, because you'll be automating something that already works instead of guessing. But the first ten? Roll up your sleeves and do it by hand. There's no shortcut, and you wouldn't want one.