How to get your startup’s first customers

Acquiring the first customers is one of the most difficult stages of building a startup. The company has no recognised brand, few or no case studies, a product that may change every week and an audience that is not necessarily looking for a business that has just appeared.

At this stage, it is tempting to believe that greater visibility will solve the problem. The founder publishes content, prepares a campaign, creates a waiting list or spends time perfecting the website. These activities may help, but they do not address the main challenge by themselves: finding a small number of people who experience the problem intensely enough, trust a new proposition and agree to commit time or money.

The first 10 customers rarely arrive through a perfectly automated channel. They come through conversations, introductions, manually delivered tests, recommendations and direct founder-led contact. This process does not scale, but it performs a function that currently matters more than scale: teaching the company who buys, why they buy, what they need in order to decide and what must happen for them to obtain value.

Finding the first customers is therefore not merely a sales task. It also forms part of validating the problem, proposition, price, product and delivery process.

First define what you will count as a customer

Before trying to acquire 10 customers, clarify what “customer” means within your model. A person who leaves an email address, participates in an interview or tests a free version may provide useful information, but has not necessarily demonstrated a willingness to buy.

A customer should make a commitment consistent with the business you are attempting to build. In most cases, that commitment will be a payment. In enterprise products with longer processes, it may initially take the form of a pilot with conditions, named owners and a defined subsequent decision. In a marketplace, you must distinguish between people supplying the offer and those completing a transaction.

It is useful to separate several signals:

• Contact: belongs to the segment and can be reached.
• Lead: has shown enough interest to continue a conversation.
• Trial user: has used the product or participated in a pilot.
• Customer: has accepted an offer and made the agreed commitment, normally a payment.
• Activated customer: has reached the first valuable outcome.
• Recurring customer: continues using, renewing or purchasing.

This distinction matters because a startup can accumulate hundreds of registrations without demonstrating that anybody will pay or remain. The first 10 customers should help you explore the complete journey from interest to value, rather than merely increasing a number at the top of the funnel.

Choose a sufficiently specific initial segment

Finding customers becomes difficult when the proposition targets “entrepreneurs”, “small businesses”, “freelancers” or “people interested in technology”. These groups are too broad to share a problem, language or purchasing process automatically.

The first segment does not need to represent the entire future market. It should allow you to find comparable people, understand their context and repeat a sales approach several times.

Instead of targeting every small business, you could begin with professional firms employing between 10 and 50 people that prepare reports manually every week. Instead of all freelancers, you might focus on independent designers managing several projects with external collaborators and deliveries. Instead of every retailer, you could choose clinics that lose appointments because customers fail to confirm.

A useful initial segment usually shares:

• A recognisable situation in which the problem appears.
• Similar consequences when the problem remains unresolved.
• Comparable current alternatives.
• A common language for describing the need.
• A reasonably accessible method of locating its members.
• An identifiable person capable of making or influencing the purchase.

The more clearly the initial segment is defined, the easier it becomes to write a relevant message, select where to find prospects and understand why a conversation advances or stops.

The route depends on the business model

Not every startup should look for its first customers in the same way. Direct founder-led selling works particularly well in B2B, services and products addressing complex needs, but the approach must reflect the type of decision involved.

In B2B

You usually need to identify both the person experiencing the problem and the person controlling the budget or authorising implementation. Early sales tend to require conversations, demonstrations, proposals and follow-up. LinkedIn, direct email, introductions and specialist events can be useful because they allow the founder to select particular accounts.

In B2C

The process may be more transactional, but you still need to find contexts in which the need already exists. Specialist communities, collaborations with small creators, events, presales or highly limited campaigns can help you observe real purchases. A large audience will not compensate for an irrelevant proposition.

In services

The professional network, recommendations and direct contact are often the fastest channels. The challenge is not merely obtaining projects, but identifying which part of the work can become a repeatable offer instead of accepting 10 completely different assignments.

In marketplaces

You do not have a single acquisition problem but at least two. You need to decide which side must be present first, how you will create value before reaching liquidity and which transactions can be facilitated manually. Ten registered suppliers do not equal 10 customers when nobody completes a transaction.

The tactic changes, but the logic remains the same: select a group, understand its decision and create a direct way to observe whether it accepts the proposition.

Create an offer that somebody can accept now

Many startups begin looking for customers when they can only describe a future vision. They explain the platform they intend to build, the features it will include or the market they want to transform, but no specific offer exists for the other person to accept.

An initial offer should explain clearly:

• Who it is for.
• Which problem or outcome it addresses.
• What exactly is included.
• What the customer must provide.
• How long the trial or service lasts.
• Which price or commitment it requires.
• What happens when it ends.

If the product is unfinished, you can offer a pilot, an assisted version or a service that produces part of the outcome manually. The important point is to describe the state of the solution honestly and establish expectations you can fulfil.

For example, a startup intending to automate reports does not need to promise a complete platform. It could offer five companies a four-week pilot in which it collects their data, generates the reports through a semi-automated process and measures the time saved. This pilot reveals issues involving data access, customisation, value and willingness to continue.

A specific offer turns an interesting conversation into a decision. Without an offer, the prospect can only express an opinion. With conditions, duration and price, they must decide whether the problem deserves a commitment.

Build a small, carefully selected prospect list

The first customers rarely appear because you publish a general invitation and wait. Build a list of people or accounts that fit the segment and display an observable reason for experiencing the problem.

The list might begin with 30, 50 or 100 prospects, depending on the model, price and ease of contact. The number is not a rule. Its purpose is to force the transition from “we sell to small businesses” to named accounts that can be researched and contacted.

For each prospect, record:

• Why the account fits the segment.
• Which signal suggests that the problem may exist.
• Who appears to use, influence or decide.
• Which connection or channel provides access to that person.
• Which message was sent.
• The response and next step.

A well-selected list generates learning even when people do not respond. If one subsector produces many responses and another produces none, the difference deserves investigation. If nobody recognises the problem, the message may be unclear or the segment may be unsuitable.

Begin with your network, but do not sell only to friends

Your professional network can reduce the trust barrier surrounding the first conversations. Former colleagues, customers, suppliers, community members and second-degree connections may be able to introduce relevant profiles.

The objective is not to ask everybody in your address book to buy. It is to use the network to reach people who genuinely belong to the segment. A friend who agrees to help because they care about you may provide feedback but does not demonstrate market demand.

A useful request is specific: “I am speaking to operations managers at small consultancies that prepare reports manually every week. Do you know anybody who fits that description and could explain how they manage the process?” This makes it easier for the other person to identify a relevant introduction.

After a valuable conversation, you can also ask whether the interviewee knows somebody with a similar context. Referrals work better when people understand exactly whom you are looking for and why the conversation may be relevant.

Use personalised direct outreach

Direct founder outreach is one of the most useful early channels because it allows you to choose whom to approach, adapt the message and hear objections without intermediaries. It may take place through LinkedIn, email, professional communities or shared contacts.

The initial message does not need to explain the complete company. It should demonstrate a specific reason for contacting that person and propose a simple next step.

A possible structure is:

• What you observed about the person or company.
• Which problem you are researching or solving.
• Why it may be relevant within their context.
• The specific next step you propose: a short conversation, demonstration or pilot.

For example:

Hi Anna, I noticed that your team manages several simultaneous projects for industrial clients. We are working with consultancies that spend a significant amount of time consolidating project updates and reports manually. I would like to understand how your team handles this work and whether the pilot we are preparing might be relevant to companies like yours. Would you be open to a 20-minute conversation next week?

The message does not need to pretend that a relationship exists or use artificial urgency. Its purpose is to secure a relevant conversation, not complete the entire sale inside a LinkedIn request.

Participate in communities without treating them as databases

Communities can help you find early customers when they bring together people sharing a professional, geographic or specialist context. Entering a group and immediately publishing a generic offer, however, rarely creates trust.

Observe which questions appear, the language people use and the solutions they recommend. Participate when you can contribute something useful and contact somebody privately only when a clear connection exists between the conversation and your proposition.

A community creates more value as a source of context and relationships than as an improvised email list. If members believe you are present only to extract contacts, you will lose the trust that makes the environment useful.

Turn the conversation into a sale, not only feedback

Early conversations often become endless interviews. The founder asks questions, listens, thanks the person and promises to send an update when the product becomes available. The company learns but never establishes whether the person would accept the offer.

Research and selling can occur within the same process without becoming aggressive. First understand the current situation: what happens, how frequently, how it is managed and which consequences it produces. When a reasonable fit exists, explain the proposition and request a specific decision.

The conversation should clarify:

• Whether the person or company genuinely experiences the problem.
• Its priority relative to other needs.
• What has already been attempted.
• Who participates in the decision.
• Which conditions are required to test or purchase.
• Which objection prevents progress.
• The next step and when it will occur.

A meeting ending with “we will speak again” provides little information. A meeting that concludes with an agreed demonstration, pending proposal, reasoned rejection or dated decision allows the founder to interpret the sales process more clearly.

Design pilots that produce evidence

Pilots can reduce the risk of adopting a new product, particularly in B2B. An unconditional free pilot, however, may attract curious users who never intended to purchase.

A useful pilot should define:

• The specific problem it is intended to solve.
• The owner on the customer’s side.
• Which information or access the customer must provide.
• Its duration.
• The result that will be measured.
• The support included.
• The decision that will be made at the end.
• Its price or the reason it is free.

Charge for the pilot whenever possible. The amount may be lower than the future price when the solution still requires collaboration and learning, but the discount should have an explicit reason. You can also offer preferential conditions in exchange for participation, structured feedback or permission to document the case.

If you decide to run a free test, limit its duration, scope and number of participants. Free access should help test a specific hypothesis rather than become the only way to generate usage.

Do not confuse a waiting list with customers

A waiting list may help measure whether a message creates curiosity, capture contacts and select people for interviews or tests. The signal is stronger when traffic comes from the intended segment and the page describes a specific proposition.

Leaving an email address still requires very little commitment. It does not demonstrate that the person will use the product, complete setup or pay. Even a large waiting list may contain people attracted by a general promise who later reject the actual conditions.

Use the waiting list as the beginning of a conversation. Ask about context, select participants, present an offer and observe how many progress towards a stronger commitment. The quality of the subsequent journey matters more than the initial number.

Charge before everything feels finished

Many founders delay charging because the product does not yet contain enough features. Willingness to pay, however, does not depend only on the degree of development. It depends on whether the outcome has value and the solution can fulfil its promise.

Charging provides information that verbal interest cannot. It forces the customer to compare the proposition with other priorities and tests whether the price, purchasing process and expectations are viable.

The initial price does not need to be permanent, but it should have a rationale. You can estimate it according to expected value, alternatives, delivery costs and the level of assistance required. Avoid a symbolic price that almost anybody would accept and that offers little information about the future model.

Avoid indefinite discounts as well. If you provide special conditions to early customers, explain what they receive, what they contribute and when the terms will be reviewed. A discount should support learning rather than hide the fact that the proposition cannot survive at its intended price.

Use a funnel even while everything remains manual

Acquiring 10 customers usually requires contacting considerably more than 10 people. Some will not respond, some will not experience the problem, some will lack authority and others will postpone the purchase.

There is no universal conversion rate. As a purely illustrative example, a list of 100 prospects might generate 30 responses, 15 qualified conversations, eight proposals and four customers. Another market may require many more accounts or convert more easily.

The important issue is to record the journey:

• Prospects identified.
• Contacts attempted.
• Responses received.
• Qualified conversations.
• Demonstrations or proposals.
• Pilots started.
• Customers acquired.
• Activated and recurring customers.

These figures reveal the constraint. If nobody responds, review the list, channel or message. If conversations occur without proposals, the problem may lack priority or you may not be asking for a next step. If proposals do not convert, investigate price, trust, purchasing process or perceived value.

A four-week plan for beginning the search

Week 1: segment, problem and offer

Define the initial segment, formulate the problem hypothesis and prepare an offer that can be accepted now. Decide what you will count as a customer and which outcome they must reach to be considered activated.

Week 2: list and initial conversations

Create the first prospect list, request introductions, send the initial messages and hold conversations focused on current behaviour. Record the language, alternatives and objections you encounter.

Week 3: proposals and pilots

Adjust the message according to what you have learned and present the offer to the strongest prospects. Define the conditions, price, scope and intended result of the pilots. Do not wait for a perfect presentation before requesting a decision.

Week 4: delivery, follow-up and review

Help the first customers reach the initial outcome, observe where they become blocked and review the complete funnel. Decide which parts of the segment, offer and sales process should remain and which require another test.

Four weeks do not guarantee 10 customers. Sales cycles can be considerably longer, particularly in B2B. The plan creates a real process and prevents the founder from spending the entire month preparing materials without presenting an offer.

Learn from early customers without building 10 different products

The first customers will provide requests, objections and different ways of using the solution. Listening does not mean automatically accepting every request.

Distinguish between:

• A need repeated across several accounts within the segment.
• A barrier preventing customers from reaching the main outcome.
• A custom requirement belonging to a single company.
• An attractive feature that does not affect the purchasing decision.
• A requirement essential for operating within the market.

If every customer receives a completely different product, the company may generate revenue without learning how to build a repeatable offer. If every request is rejected to protect the original vision, the founder may ignore the information that should change it.

The role of early customers is not to design the product for you. It is to reveal which parts of the proposition repeat, where value appears and what it costs to deliver the outcome consistently.

What to measure during the first sales

The number of customers matters, but it is not enough to interpret progress. Also record:

• The percentage of prospects responding.
• How many conversations genuinely belong to the segment.
• The percentage progressing to a proposal.
• The objections appearing most frequently.
• The length of the sales cycle.
• The people involved in the decision.
• The number of pilots converting into customers.
• The manual work required by each account.
• Time to first value.
• The number continuing, renewing or referring others.

These metrics do not require a sophisticated system. A spreadsheet may be sufficient during the early stage. The important point is to preserve the history so that patterns can be distinguished from isolated impressions.

When to begin scaling acquisition

You do not need to wait for perfect product-market fit, but you do need some repetition. Before investing significantly in campaigns, automation or a sales team, you should understand reasonably well:

• Which segment responds best.
• Which problem causes the purchase.
• Who makes the decision.
• Which message opens conversations.
• Which objections block progress.
• Which journey leads to first value.
• Which parts of delivery still depend on the founder.
• Whether the price can support the cost of acquiring and serving the customer.

Scaling a process that does not work will not repair it. It will allow the problem to consume budget more quickly. The first customers exist partly to build a process worth repeating.

How Foundeia organises this process

Customer acquisition should not be treated as an isolated marketing task. The selected segment affects the proposition; the proposition determines which offer can be made; conversations reveal objections; pilots expose the cost of delivery; and payments provide information about the business model.

Foundeia connects these decisions so that the founder does not accumulate contacts without learning why some progress and others do not. Artificial intelligence can help prepare messages, organise notes, compare segments and identify patterns, but the evidence comes from market behaviour.

The objective is not to automate outreach immediately or generate hundreds of messages. It is to build a coherent sequence: select, contact, understand, offer, deliver, measure and decide.

Once this journey is recorded, early customers stop being isolated cases. They begin to reveal which parts of the sale can be repeated and what must change before attempting to acquire the next hundred.

The first 10 customers are the beginning of the evidence

Acquiring 10 customers does not automatically validate the entire company. It may demonstrate that particular people accept an offer under specific conditions, while leaving questions about recurrence, retention, margin, channels and delivery capacity unanswered.

The value lies in the quality of the learning they produce. Ten customers from the same segment, attracted by a similar problem and capable of achieving a comparable outcome provide more information than 10 unrelated sales secured through personal relationships or exceptional discounts.

You do not need to find a scalable channel immediately. You need to discover whom you can help, which proposition they will accept, what it costs to create the outcome and which part of the process can be repeated.

Early customers appear when the startup stops waiting for an abstract audience and begins holding specific conversations. Scale can follow. First, you need a sale you can explain, deliver and attempt again.

Frequently asked questions