How to Get Your First 100 Customers

The hardest customers a startup ever gets are the first ones. You have no brand, no reviews, no case studies, and no proof anyone else trusts you. Every later stage of growth has some momentum to borrow from; the first hundred customers have none. You earn them one painful conversation at a time.

This is also the stage where founders reach for the wrong tools. They build a marketing funnel, run some ads, write a few blog posts, and wait. Nothing happens, because scalable marketing is designed to amplify a message that already works on a customer you already understand — and at the start you have neither. The first customers come from somewhere else entirely.

This guide covers where early customers actually come from, why unscalable tactics beat marketing machinery at the start, and a channel-by-channel playbook for getting to your first hundred.

Do things that don't scale

The most useful advice on early customers is Paul Graham's: do things that don't scale. The instinct to automate and systematize from day one is exactly backwards. At the start, the manual, unscalable, embarrassingly hands-on approach is not a compromise — it is the only thing that works, and it is how every later channel gets discovered.

Recruit customers one at a time. Email people you know have the problem. Show up in person. Onboard each user yourself and watch where they struggle. Give early customers a level of attention you could never sustain with a thousand of them. This does two things at once: it gets you the customer, and it teaches you who they are and why they said yes — the knowledge that everything scalable is later built on.

The most common unscalable thing founders have to do at the start is to recruit users manually. Nearly all startups have to.

Go where your customers already are

You do not need to invent an audience. For almost any problem, the people who have it already gather somewhere — a subreddit, a Slack or Discord community, a LinkedIn niche, a local meetup, a forum, an industry event. Your job is to find those rooms and become a genuinely useful presence in them before you ever pitch.

The mistake is broadcasting. Posting "check out my product" into a community gets ignored or banned. Answering questions, sharing what you have learned, and reaching out privately to people who describe the exact problem you solve gets you customers. It is the same principle that gets a founder a co-founder through the communities where builders already gather: contribute first, and the asks land.

14%

A channel-by-channel starting playbook

Different products reach their first customers through different doors. Start with the one closest to your buyer.

None of these is a growth engine yet — they are ways to start conversations. The point is to get enough real customers that a pattern emerges: which type of person buys, which message lands, which channel reliably produces a yes. That pattern is the prize.

Treat every early customer as a research interview

The reason the first hundred customers matter more than the next thousand is that each one is a free, high-resolution research interview. Every conversation tells you who your real customer is (often not who you guessed), what language they use for the problem, what almost stopped them from buying, and what made them say yes.

This is the same evidence you started gathering when you validated the idea — now sharpened by people actually paying. Capture it. After every early sale and every loss, write down why. Within a few dozen customers you will see the repeatable story: the specific person, the specific reason, the specific channel. That story is what later marketing amplifies, and it is also the clearest evidence you are approaching product-market fit.

Your first-100 plan

• Customers 1–10: Pure manual recruiting. Founder-led outreach to people you know have the problem. Onboard each by hand, learn obsessively.
• Customers 10–50: Double down on whatever produced the first ten. Get deep in the two communities where your buyers gather. Start noticing the pattern.
• Customers 50–100: Name the pattern — who, why, where — and test whether one channel can be made repeatable. Only now does it make sense to consider paying to scale it.

Getting the first hundred customers is slow, unglamorous, and impossible to automate, which is exactly why it filters out the founders who were never going to make it. The ones who win treat it not as a marketing problem to outsource but as the most important learning of the company's life. If your product helps other founders or startups, the Foundersbase network is one place your earliest customers are already looking.