The First 10 Customers: How Founders Actually Get Them

Most founders confuse the first ten customers with the first thousand. They are not the same problem. Customers one through ten are a trust transfer. Someone you barely know agrees to hand you money, attention, or a seat at the table before you have proof you deserve it. That only happens when you ask a specific person for a specific thing in a specific way.

Paul Graham has written that the top reason early startups die is a failure to 'do things that don't scale' [1]. The first ten customers are the clearest test of that idea. You are not optimizing a funnel. You are knocking on doors. The sooner you accept that, the sooner you start getting to yes.

There is another reason this stage is unique. At zero customers you have total storytelling freedom. By the time you reach a hundred, your positioning is locked in by the people already paying you. The first ten get to decide who you serve for the next two years. Pick them on purpose.

Before you build a landing page or cold email list, do this exercise on paper. Write a single sentence that describes one real human being who would pay you this week if you could reach them.

Bad version: 'Small business owners who want more leads.'

Good version: 'Solo physical therapy clinic owners in Austin with one or two locations, who still do their own scheduling, and who lost a patient last month because they could not return a call in time.'

The specificity is not decoration. It tells you which Slack groups to join, which associations to call, which LinkedIn filters to apply, and which objection to prepare for. Founders who skip this step almost always end up pitching a general audience with a general message, which gets general indifference.

First Round Review calls this 'the minimum viable segment' and their interviews with early stage founders consistently show that narrowing the audience is the single biggest unlock before traction [2].

Your first ten customers will come from a list of roughly 100 prospects. That ratio is remarkably consistent across SaaS, services, marketplaces, and physical products in the zero to one stage. Plan for it.

Put 100 names in a spreadsheet with four columns: name, role, how you can reach them, and what you know about them personally. Sources for the list include:

• Your phone contacts and LinkedIn connections
• Customers of the five most similar products or services to yours
• Members of associations, meetups, subreddits, or Discord communities where your target spends time
• Commenters on articles or YouTube videos about the problem you solve
• People who have publicly complained about the problem on X or Reddit in the last 90 days

If you cannot reach 100 names, your segment is too narrow or you have not actually talked to anyone in it. Go back to step one.

Every message in the first batch is written one at a time. No templates, no mail merge, no sequencer. Yes, it is slow. That is the point. You are collecting evidence about what resonates before you scale anything.

A message that consistently earns replies from cold prospects at this stage has four parts:

1. A sentence that proves you are not a bot. Mention something specific about them, their company, or a post they made.
2. The exact problem you think they have, phrased the way they would phrase it.
3. What you are building to solve it, in plain language, with no jargon.
4. A small ask. Fifteen minutes on a call, or 'could I send you the early version and get five minutes of your honest reaction?'

Do not ask for the sale in message one. Ask for the conversation. The sale happens in the conversation when you discover what they actually need.

On the first call, stay in discovery for most of it. Then, in the last five minutes, ask three questions in order:

1. 'Based on what you have described, does this sound like something that could help you?'
2. 'If we could get it working for you by [specific date], what would need to be true?'
3. 'What would stop you from trying it?'

Question one tests fit. Question two pulls the customer into co-owning the outcome. Question three surfaces the real objection, which is almost never price and almost always risk. Address the real objection and you will close more often than not.

Many founders never ask these questions because they are afraid of the answer. The answer is the whole reason you took the call. Ask.

Your first ten customers should get a level of service that is obviously unsustainable. That is not a failure. It is a feature. You are buying two things that cannot be purchased with money: raw insight into what actually matters, and customer stories you can use for the next hundred prospects.

Concretely, that means:

• A private Slack, WhatsApp, or text thread with each of them
• A personal check-in every week or two
• Fast turnarounds on feature requests they care about
• A written case study or testimonial once you have delivered real value

Every hour you spend here compounds. Your next ten customers will come from referrals, screenshots, and quoted outcomes from the first ten. Treat the first ten like founders. They are betting on you.

Three failure patterns kill first ten customer efforts more than anything else.

First, building before selling. If you spent the last six weeks on a website and have not had ten real conversations, you are avoiding the work that matters. Stop and talk to people today.

Second, running ads too early. Paid acquisition is a scaling mechanism. Until you can reliably close a conversation, paying for more conversations just teaches you to lose faster.

Third, pricing apologetically. Low prices do not make it easier to close. They often make it harder because they signal you do not believe in the value. Pick a price that reflects the problem's cost, then charge it.

How long should it take to get the first 10 customers?

For most B2B and service businesses, four to twelve weeks of concentrated outreach is realistic once you have your segment defined. Consumer products often take longer because the average order value is smaller and individual conversations are less efficient.

What if no one replies to my messages?

Almost always, the message is too long, too generic, or the segment is wrong. Cut the message in half, replace every sentence that could apply to 'any company' with something specific, and reconfirm that your list is actually the people who have the problem.

Should I offer a free trial or discount?

A time boxed pilot at full price almost always beats a free trial. Paying customers give better feedback, show up to meetings, and convert at dramatically higher rates than free users.

How do I find customers if I do not have a network?

Buy one. Spend two weeks joining the three communities where your target hangs out, contributing real value with no pitch, and building relationships. Your network is the project, not the obstacle.

1. Do Things That Don't Scale - Paul Graham
2. First Round Review on Minimum Viable Segment
3. Y Combinator: How to Get Your First Ten Customers
4. Lenny's Newsletter: How the Best Founders Find Their First Customers