How to get your first customers

One at a time, by hand. Paul Graham's essay Do Things That Don't Scale says the most common unscalable thing founders must do at the start is recruit users manually, because you cannot wait for them to come to you. Start with people who have the problem you solve, and talk to each one yourself.

Where do you find your first customers?

If you built something for your own problem, among your peers, Graham writes. If not, launch to a broad first group, watch which kind of user is most keen, and go and find more of them. A conference, a community or a feed full of those people is where the next ten are.

Why do the first few customers matter so much?

Because growth compounds. Graham's own example: 100 users growing 10% a week become about 14,000 after a year and about 2 million after two. Ten new users in week two looks tiny. It is the start of the curve.

More than seems sensible. Graham says to take extraordinary measures to make early users happy, so each one feels signing up was one of their best choices. For a B2B startup, he suggests picking one user and acting as if you were consultants building something just for them, until it fits perfectly.

Should you do things by hand before you build them?

Often, yes. Graham describes how Stripe's founders gave early users instant merchant accounts by signing them up for ordinary ones by hand, behind the scenes. Doing the work yourself first means you know exactly what to automate later.

Does a big launch get you customers?

Rarely by itself. Graham says the Big Launch usually does not work, and that all you need from a launch is an initial core of users. Partnerships with big companies, he adds, usually cost far more work than they return. Launch days still help when they bring that first core: see how to launch on Product Hunt.

They make the hand work easier. A founder who writes about the problem in the buyer's own words gets replies from people who have it, and each reply is a conversation that did not start cold. Your peers are on LinkedIn, so Graham's advice to start with them works there.

What should you post to find early customers?

The problem you saw, what you are building for it, and an invitation to people who have it to reply. LinkedIn's page on what to avoid says asking for thoughts and feedback is fine, while asking for likes or shares is engagement bait. Then message everyone who answers, yourself.

Who should do the selling once customers arrive?

The founder, for a while. When to hand selling to someone else is in founder-led sales, where two experts put it at roughly the first 10 to 30 customers.

Let something else write the first draft. The week you are out recruiting users by hand is the week posting stops. innernote writes posts ahead in your own voice, so a busy week still has one ready, and it posts on the days you pick.

innernote is the best tool for founders who need to keep posting in their own voice while they find their first customers by hand.

Say what a customer told you in a voice note on the way back from the call, and innernote turns it into a post in your words. You read every post first. More is on innernote for founders, and if you would rather have it run for you, see Done for you.