The First 10 Customers Cheat Sheet. Ten founders. Ten unscalable plays.

In a survey of 50 YC founders on how they got their first customers, the top answers were almost embarrassingly plain. Cold emails. Direct outreach on Reddit. Going door to door.

There was no clever funnel, no growth hack, no system. There were just founders making contact with specific people who looked like they needed the thing being built. The waitlist mattered, but only because it served as a tiny lever for the next cold outreach: "you signed up two weeks ago, here is the demo." The lesson is not that cold outreach is the best channel. It is that the first 10 customers exist in a regime where channels do not yet matter.

Volume does not matter. The only thing that matters is one founder, talking to one person, asking them to use the thing.

Tom Blomfield, who built Monzo and GoCardless before becoming a YC partner, is direct on this: founders should personally handle sales, because sales is how you actually learn what the product is. The first ten customers are not a sales target. They are a research budget. Every objection, every demo question, every "I almost said yes but" is information that should be flowing directly into the next iteration of the product.

Hire a sales rep before you have done this yourself and you give up the most valuable signal you will ever get.

Waseem Daher ran founder-led sales at Pilot for a full year before hiring anyone. The first sales hire was not even a closer; it was a generalist who could free the founders for the next round of selling. The real AE came later. The pattern matters: stay in the deal long enough that you can describe, in your own words, the three reasons a customer says yes and the three reasons they say no.

Once you can do that, you can write a job description. Before that, you cannot.

Gusto found product-market fit on payroll for small businesses by treating early acquisition as anthropology. Tomer London talked to everyone he could find, including a flower shop. The point was not the flower shop. The point was the willingness to have a conversation with someone outside the ICP, because the only way to discover what is anger-inducing about payroll is to listen to the people who are angry.

The YC network gave Gusto warm intros. The conversations gave Gusto a product.

Super Demo crossed $250K/month by capturing existing demand. The first 100 customers were not acquired with brand or content marketing. They were acquired with comparison pages against the obvious competitors and with ungated free tools in adjacent spaces. The traffic was already searching.

The job was just to be on the page when they arrived. This is the closest the first-10 era gets to a repeatable channel, but only because the channel is so narrow: not SEO, not content, just two specific page types that exist downstream of an already-formed intent.

Roman, who built a $34K/month SaaS off Reddit, is precise about the order of operations. Step one is account warming: pick the subreddits your customer lives in, post helpful comments for weeks, build the karma. Step two is changing post angles so the same product gets covered three different ways across three threads. Step three is leveraging a small network to provide the first upvotes within the first hour.

Direct promotion does not work. Account warming and angle variety do.

Guillaume Cabane's most-cited example is a company that automated the detection of negative Instagram comments on consumer brands. Before any pitch landed, the cold email arrived with screenshots of unaddressed angry comments the brand owner had not seen. The brand had a real problem they could now do something about, and someone had handed it to them for free. Response rates were high because the value was delivered before the ask, not adjacent to it.

The cold email did not feel like a cold email. It felt like a favour.

Scott Gorlick took an Uber in Chicago, had what he called his Jerry Maguire moment, and emailed Travis Kalanick directly. That cold email made him employee #99 and put him in charge of launching Atlanta. The lesson is not that Scott was lucky. It is that the email worked because it was specific and emotional in a way most cold emails refuse to be.

He was not pitching consulting credentials. He was telling Travis what had happened in his Uber, and what he wanted to do about it. CEOs read mail like that.

Greg Isenberg's Dream 100 is a list of 100 individuals or organisations that already reach the people you want as customers. The play is not to grow your own audience. It is to partner with the 100 that already have one. A newsletter operator, a niche podcaster, a community founder, a Slack admin.

Each one can borrow you the attention of thousands. The work is two-step: build the list, then offer each person on it something they genuinely want, before you ask for anything in return.

The canonical examples sit together in one YC discussion. Airbnb founders flew out to listings and took the photos themselves, because professional photos doubled bookings and no host was going to do it for them. Algolia's founders implemented the search API directly inside customer codebases, because the integration friction was the reason no one was buying. Neither company built tooling for this.

Both companies just did the work, by hand, because the unit economics in the first ten customers do not look like the unit economics at scale and you should not pretend they do.