Your first customers are not simply a smaller version of a future marketing funnel. They are people making a particular buying decision about an offer that may still be changing. Reaching them requires a clear promise, a credible way to deliver it, and an honest understanding of why they would consider switching from what they do now.
The “first ten” in this guide is a manageable planning frame, not a guaranteed outcome or a scientifically validated threshold. We will use a hypothetical service that organizes product photographs for independent online shops. The examples are original exercises, not reported campaign results or advice to send unsolicited bulk messages.
Narrow the offer before widening the audience
State the customer and the job in one sentence. “We help businesses with content” is hard to evaluate. “We organize a small shop's existing product photographs into a consistent, named catalog” describes an output a buyer can understand. It also makes exclusions easier to explain.
Choose an initial group whose workflow you can learn. For the example, that might be owner-operated shops with an existing product range and disorganized image files. It would not automatically include large retailers, photographers seeking new clients, or businesses that need images created from scratch.
Write the scope, delivery time, inputs, revision boundary, and price basis. When these are vague, a promising sales conversation can turn into an obligation the founder cannot deliver. Use the pricing and unit economics guide to check the work behind the promise before expanding outreach.
Build a small, relevant prospect list
Start with organizations for which you can explain a plausible connection between the observed workflow and your offer. Use appropriate public business information or introductions, and respect platform rules and applicable communication requirements. Do not collect personal data simply because it is technically accessible.
For each prospect, record why the offer may be relevant and what you still do not know. Avoid treating an imperfect website as proof that the owner is dissatisfied or can afford a new service. Observations create questions; they do not reveal private intentions.
Keep the list small enough to research carefully. The purpose is not to maximize a contact count. It is to make a limited number of respectful, specific approaches and learn whether the problem and offer match. A larger list will not repair a message that is irrelevant to its recipients.
Make the first message concrete and honest
A research invitation, not a sales claim
A useful first message identifies you, explains the reason for contact, and makes one modest request. For example: “I work on product-image organization for small shops. I am researching how owners manage image files across listings. Would a brief conversation about your current process be relevant?” Adapt the wording to the actual relationship and communication rules.
Do not pretend to have audited a business when you have only looked at one public page. Do not claim a mutual contact who has not made an introduction. Avoid invented scarcity, guaranteed sales improvements, or language implying that the recipient has already expressed interest.
Where appropriate, make declining easy and respect it. A non-response is not permission for an unlimited sequence. Keep your communication policy conservative and suitable for the jurisdiction and channel. The goal is a credible business conversation, not forcing attention through persistence.
Learn before proposing a pilot
Ask how the customer currently organizes photographs, who performs the work, and what happens when a file is missing or inconsistent. Do not assume that the visible symptom is the task they most want to change. Their actual priority may be creating new images rather than reorganizing old ones.
Y Combinator's essential startup advice emphasizes talking to users and doing early work that may not scale. Here, that principle is applied as a suggestion to learn the workflow directly before trying to automate acquisition. It is not evidence that this specific service or outreach approach will succeed.
Separate research from a sales proposal. If the conversation reveals a suitable need, ask whether the person would like to discuss a defined pilot. The customer interview guide offers a structure for understanding the task without making every question a disguised pitch.
Design a pilot you can finish well
A pilot should specify the material to be processed, the output, delivery conditions, responsibilities, and the way payment works. For the hypothetical catalog service, that might mean organizing an agreed folder of authorized images into a stated naming scheme. It should not quietly expand into new photography, copywriting, and unlimited listing management.
Explain what is manual and what is automated. Early hands-on delivery can be useful, but it should not be misrepresented as a finished software platform. Do not make security or reliability promises that your actual process cannot support.
Set an acceptance process before starting. Ask how the customer will confirm that the agreed output is usable and what happens if a file is missing. When permissions, intellectual property, or other legal issues are material, obtain appropriate advice and written agreements rather than assuming a friendly conversation resolves them.
Track the path from conversation to completed work
Keep stages distinct: contacted, replied, qualified conversation, proposal, accepted engagement, delivered work, and collected payment. An enthusiastic reply is not the same as a customer. A signed agreement is not necessarily collected revenue. Clear definitions keep the review from becoming a collection of flattering numbers.
Record why an opportunity moved or stopped. A buyer may lack the need, prefer another solution, face a timing constraint, or find the offer unclear. Do not compress these into one “not interested” label when they suggest different next actions.
Use any rates cautiously when the sample is small. A few conversations can help identify possible problems with the offer, but they do not establish a stable conversion benchmark. Keep individual notes and context alongside totals so that the arithmetic does not appear more informative than the evidence.
Ask for feedback before asking for referrals
After delivery, ask what was useful, what was confusing, and which part required more work from the customer than expected. Compare that feedback with the original promise. A completed project can still reveal a scope or process problem that should be fixed before another sale.
When the customer is comfortable and a referral is appropriate, ask whether they know another business facing the same task. Do not imply that a referral is required, and do not use the customer's name as an endorsement without permission. An introduction and a public testimonial are different permissions.
Avoid turning one satisfied customer into a broad performance claim. You can describe the specific work completed when authorized, but claims about revenue improvements, time savings, or outcomes need evidence. A carefully documented small project is more credible than a generalized success story built around it.
Decide what should become repeatable
After several engagements, examine which steps recur and which depend on exceptional effort. If every project requires a different service, the offer may need a narrower boundary. If the same misunderstanding appears repeatedly, revise the explanation or handover rather than blaming every customer.
Identify the next constraint before choosing a channel to scale. It may be delivery capacity, unclear scope, insufficient proof, or a mismatch between the buyer and the user. More traffic will not necessarily resolve those problems and may make them more expensive.
The Start a Business track connects this work to validation and launch planning. Treat early customer acquisition as a learning process with real delivery responsibilities, not merely a race to reach a public milestone.
Conclusion: earn a repeatable next conversation
Finding the first ten customers starts with a specific offer and a reason to approach a specific buyer. Research respectfully, listen before proposing, deliver within a clear scope, and distinguish interest from completed sales. The useful outcome is not only a customer count. It is a better-supported understanding of who the service fits and what the business can reliably promise next.
