From idea to your first ten customers: a starter playbook

Tactics for ten customers and tactics for ten thousand are different jobs. Most early founders learn this expensively, by trying to scale before they have anything that actually works for one person. This lesson is the simple version of going from idea to your first ten paying customers: the steps in the right order, what to ignore, and the moment you know the early game is over.

The first ten is not a smaller version of the next thousand. It is a different shape of work entirely. You are not optimizing a funnel; you do not have one. You are not picking between channels; you do not have data yet. What you are doing is finding out whether anything you believe is actually true. Speed and honesty matter more than polish.

// 01Why the first ten is a different game

Three things are unusual about the early stage that change what is worth doing.

• You do not have statistics yet: With three users, every percentage is meaningless. The signal you can read is qualitative: did they keep using it, did they pay, did they tell anyone, did they look surprised when something worked.
• You can do things that do not scale: Onboard each user by hand. Send personal emails. Watch them use the product over a screenshare. None of this works at a thousand customers; all of it works at ten and teaches you everything.
• You can change the product fast: Nothing is locked in yet. There is no migration to consider, no enterprise customer to placate. The cost of a wrong feature is small; the cost of building five wrong features in a row is large.

// 02Step 1: name a real problem and a real customer

Before any code or copy, write the two paragraphs from lesson 2: who the customer is and what they are trying to get done. Make it specific enough that you can name three real people who fit. If you cannot, you do not have a customer; you have a wish.

• Avoid the word "everyone": and any synonym for it. The narrower the description, the easier every later decision becomes.
• Pick a problem you can describe in one sentence: and that sentence has to use the customer’s words, not yours. If your sentence has the word "platform" or "solution" in it, rewrite.

// 03Step 2: build the smallest version that proves something

Most founders build too much for the first version. The right question is not “what is the minimum feature set,” it is “what is the smallest thing that could prove someone will pay for this.”

• Sometimes the answer is a landing page: A page describing the product, with a button that says "Pay now" or "Get on the waitlist." If nobody clicks, you do not need to build anything; if people click, the next step is clear.
• Sometimes the answer is a service: You do the work manually for the first ten customers, with no automation at all. You are testing whether the result is valuable; the way you produce it can be embarrassing.
• Sometimes the answer is a tiny product: One feature, one screen, one workflow. Built quickly and shipped to actual users within two weeks. Slow shipping is the enemy at this stage.

Whichever shape you pick, the test is the same: did someone do something costly (pay, sign up, refer a friend, respond with a paragraph of detail) on the basis of what you showed them? That is the signal. Vague approval is not.

// 04Step 3: do things that do not scale

At ten users, your job is to be a remarkable concierge. Personally onboard each one. Be present in their first session. Answer support requests within an hour. Send a thank-you note when they pay. None of this works when you have a thousand users; all of it is the single most valuable thing you can do at the start.

• You learn what is broken in real time: The user gets confused, you fix it before they leave. You hear five users say the same thing, you change the homepage that afternoon.
• You build durable relationships: Customer #3 who got personal attention will be customer #3 who refers customer #15. The work compounds.
• You stay grounded: Founders who automate too early lose touch with what their customers are actually feeling. The instinct of "I should know what users are running into right now" never gets developed.

// 05Step 4: charge real money or get a real commitment

Free is not a price; it is the absence of a price. Free users tell you nothing about whether the product is worth money to them. The earlier you ask people to pay, even a small amount, the earlier you find out whether the value is real.

• Charge less than you think you should: Early prices are a tool for learning, not for revenue. The number can move later; for now you want as little friction as possible while still asking for money.
• A non-cash commitment can also work: A signed letter of intent, a public testimonial, a calendar invite to a paid pilot. Anything costly that signals "yes, I will commit something." Words alone are not enough.
• Treat refunds as data: A customer who churns and asks for a refund is teaching you something specific. Take their feedback seriously. They are the most honest user you have.

// 06Step 5: write down what you learned

After the first ten customers, the most valuable thing you can do is sit down and write, in plain language, what is now true that was not true before.

• Who actually fits: The customer description from lesson 2, but rewritten with everything you know now. It will be more specific.
• What they actually paid for: Not the features in your spec, but the part of the value that the dollars were really for. Often it is one specific outcome out of five.
• What did not matter: Features you built that nobody asked about, parts of the homepage that nobody read, channels that produced no signal. Cutting things is as important as adding them.
• What surprised you: The single most valuable section of the document. Surprises are the things you cannot afford to lose.

// 07What to ignore at this stage

Most of the advice on the internet aimed at startups is for companies that are already past the first ten. It is not wrong; it is just for a different problem than the one you have. The list of things to defer:

• A formal brand identity: A logo and a font choice are not what is keeping you from your eleventh customer.
• A tech stack debate: Whatever ships fastest is the right answer for now. Refactor when you have product-market fit, not before.
• Hiring: A first hire too early is an expensive way to slow yourself down. The work of the first ten customers does not parallelize well.
• Multi-channel marketing: You only need one source of users at this stage. Pick the most direct one and exhaust it before adding a second.

// 08Five things to carry forward

• 01: The first ten customers is a different job from scaling. The work is qualitative, hands-on, and faster than it looks like it should be.
• 02: The smallest version is whatever can prove someone will pay. A landing page, a manual service, a one-feature product; whichever can fail honestly fastest.
• 03: Do things that do not scale. Onboard each user by hand. The instinct you build is more valuable than the time you save by automating.
• 04: Charge real money, even a small amount, as early as you can. Money turns words into evidence.
• 05: After ten, write down what is now true. The document is the bridge from "I had an idea" to "I know what to do next."

At this point you know the customer, the approach, the product loop, the conversation, and the early playbook. The last lesson is the one that changes everything once you have the basics. It is about why opinion-driven growth fails, what an experiment actually is, and how every later decision in your career becomes a question you can answer instead of a guess you can defend.