Your First Ten Customers Won’t Come From a Tool

Y Combinator Startup School · Max, Visiting Partner at Y Combinator · Video ID: _FBivfgOvuE

Published June 22, 2026

The fastest path to early revenue rarely begins with software automation. It begins with the founder doing the uncomfortable, high-context, unmistakably human work of finding where buyers already spend time, earning trust, and showing up before any scalable channel is ready to scale.

Who Is Max?

Max is a visiting partner at Y Combinator, where he works closely with early-stage founders wrestling with founder sales, customer discovery, and the messy transition from “we know our target customer” to “we can reliably start conversations with them.” His advice is grounded in dozens of YC founder stories collected from Bookface, YC’s internal founder network, about how startups actually landed their first 10 customers.

The Central Question: Where Does Your Buyer Actually Spend Time?

Most founders default to cold email and LinkedIn because those channels are convenient. They can be operated from a laptop, instrumented with tools, and made to feel like progress. For some markets, that is rational: sales leaders live in email and LinkedIn, so sales software can often start there.

But many buyers do not organize their work lives around an inbox. A school district administrator, property manager, insurance agent, or truck dispatcher may be reachable through conferences, phone calls, job sites, Facebook groups, trade associations, or peer recommendations long before they are reachable through a polished sequence.

One YC founder spent months sending email and LinkedIn outreach into a legacy industry with terrible open and reply rates. When he finally walked the floor at an industry trade show, he closed more in three days than he had in three months of cold email.

The first sales channel decision is therefore not “which tool should we buy?” It is a customer anthropology question: What does the buyer’s day look like? How often do they check email? Which conferences do they attend? Do they answer the phone? Are they on Reddit, LinkedIn, Facebook, Discord, YouTube comments, industry forums, or trade association boards? If those answers are vague, the problem is not tooling. The problem is insufficient time with real customers.

Customers One Through Three Come From Trust, Not Targeting

The first two or three customers, almost without exception, come from people who already have some reason to trust the founder: former colleagues, classmates, friends in the industry, or people one introduction away. Early customers are not only buying product quality. They are taking a risk on the person building it.

That makes warm networks disproportionately valuable. Founders should work through personal contacts first, then second-degree LinkedIn connections, then broader network-search tools. Max mentions Happenstance, a YC company that lets founders search an extended network in natural language — for example, “find people who work on notification systems at big companies.”

The highest-leverage intro request is specific: name the person you want to meet, explain why that person would care, and write the forwardable blurb yourself. The easier the ask is to execute, the more likely the introduction happens.

Prospecting tools become useful later. Spending weeks configuring outreach software while second-degree relationships remain untouched is a classic form of premature scaling. Before 10 to 20 quality customers, the warm network is usually the lowest-hanging fruit.

Being in the Same Room Still Beats Almost Everything

YC founder stories repeatedly converge on one uncomfortable tactic: physically showing up. It is slower than software, more awkward than a Zoom call, sometimes expensive, and exposes the founder to real-time rejection. It also works.

One founder flew to meet a single executive buyer four weeks in a row. The buyer kept rescheduling; the founder kept showing up and eventually closed the account. Another founder repeatedly appeared at customer offices uninvited, was asked to leave most of the time, and even flew to Hawaii for a customer who kicked him out after eight minutes. Through persistence, that customer became one of his biggest accounts.

Small, industry-specific conferences are especially potent because one real conversation can convert better than dozens of cold emails. The tactical playbook is simple:

• Create a Calendly with 15-minute slots stacked across the conference days.
• Email the attendee list before the event and fill as many slots as possible.
• Email again during the event to catch people who ignored the first messages.
• Spend the event in back-to-back buyer conversations, not passive booth theater.

Micro-events can be even stronger. A dinner or happy hour for six to 10 ideal customers at $50 to $100 per head often converts better than large-format events. Once someone has shared a meal with a founder, ignoring the next email becomes much less likely.

Find the Public Pain and Become a Real Person There

For consumer products and many SMB markets, future customers are often already complaining in public. The founder’s job is to find those places and participate as a real person rather than as a distribution bot.

Reddit came up repeatedly. One founder posted a product demo, watched commenters tear it apart, and still attracted hundreds of quiet lurkers who signed up. Another found old threads where people complained about the exact problem he solved and messaged every commenter one by one. A healthcare founder spent months responding to Reddit and Facebook complaints, posting two to five times per day, getting shadowbanned in some subreddits, and still acquiring customers.

The principle extends beyond Reddit: Facebook groups, Discord servers, YouTube comments, industry forums, and trade association message boards can all be the place where pain is expressed. Reddit has one extra advantage: threads persist in Google search for years, so helpful responses can keep compounding long after the original post.

Outbound Starts After the Founder Has Something Worth Scaling

Once the warm network, in-person channels, and customer communities have produced learning, outbound can become useful. The founder’s job is to identify companies matching the ideal customer profile, find the right person, get contact information, and reach out with a message grounded in actual insight.

Max highlights three common tools:

The framing often matters more than the pitch. Some founders opened doors by asking for advice, mentorship, product feedback, or a whiteboard session. This only works when the learning intent is real. One founder asked CEOs in his space if they would mentor him; a few became customers. Another spoke with 200 salespeople before building the product, maxing out weekly LinkedIn connections around one hypothesis at a time. Her connection requests had about a 50% acceptance rate, and 20% of those accepted requests converted into calls.

A DevTools founder offered free whiteboarding sessions on agent architecture, then opened shared Slack channels to help teams implement the architecture — which naturally required his product. A founder selling to lawyers offered $100 to $200 per hour for product feedback, roughly a portion of their hourly fee. About 30% accepted, and the conversion rate made the acquisition cost reasonable for a high-contract-value market.

The Outreach Copy Rule: Human, Short, and Specific

The exact words matter less than founders think, but three rules matter a lot. Keep the message under 75 words. Make the call to action painfully clear. Read the message out loud to a friend and remove anything that sounds like something you would never say to a real person.

The best early outreach often gives before it asks. An API security company can run a quick public-site vulnerability scan. A mobile onboarding tool can review a prospect’s app and send specific suggestions. A compliance startup can prepare a short audit note tailored to a prospect’s product, then ask for a meeting to go deeper.

Twenty minutes of work before asking for 30 minutes of someone’s time is a reasonable trade when the goal is not scale but the first 10 customers.

Founders should also follow up three or four times over a couple of weeks. Silence is not always rejection; it is often just low priority.

The Customer Sequencing Framework

Max’s most useful frame is a staged view of early customer acquisition:

The messy middle works because the founder is personally doing the work. A researched email, a Reddit DM, or an office visit signals commitment that automation cannot fake: the founder cares enough about the problem to spend their own time on it. At the earliest stage, that is the startup’s advantage.

Key Lessons

• Start with buyer habitat, not channel fashion. If the buyer does not live in email, email optimization is wasted motion.
• Use trust before tools. Customers one through three usually come from people close enough to bet on the founder.
• Prefer proximity when stakes are high. Conferences, office visits, dinners, and workshops create trust faster than sequences.
• Participate where pain is already public. Reddit, Facebook groups, Discord, YouTube comments, and forums are discovery surfaces, not just marketing channels.
• Do manual work before automation. Tools should scale a proven message, not substitute for learning what message works.

Why This Matters for Diffie

For Anand and Diffie, the immediate question is not “Which outbound stack should we run?” but “Where do frontend engineers, QA leads, and engineering managers already reveal pain around browser testing?” The likely habitats are narrower than generic startup LinkedIn: frontend communities, Playwright and Cypress discussions, flaky-test threads, QA Slack/Discord groups, GitHub issues, engineering-manager peer groups, and teams visibly shipping complex frontends.

Customers one through three should come from trust: ex-YC founders, former colleagues, engineering leaders in Anand’s network, and second-degree intros to teams with obvious frontend QA pain. The ask should be specific and low-friction: “Can I watch how your team currently catches frontend regressions and show you where Diffie would have flagged them?” That is stronger than a generic AI browser-testing pitch.

For customers four through 10, Diffie should lean into unscalable founder-led proof. Offer free teardown sessions of a prospect’s web app or CI test suite. Record a short audit showing missed visual, interaction, or browser-flow regressions. Join the communities where engineers complain about flaky tests and brittle E2E suites, and respond with concrete diagnosis rather than promotion. For high-value prospects, a 20-minute custom browser-testing audit before asking for a 30-minute call is exactly the kind of founder-led work that can earn attention.

Only after those conversations produce repeated language — the same failure modes, the same buyer objections, the same “we would pay if…” pattern — should Diffie scale with Apollo, Clay, or sequences. The message worth scaling will not be “AI browser testing.” It will be the sentence Anand hears from real teams often enough to recognize as demand.