First Customers Come From Founder-Led Work, Not Sales Automation

Max Kolysh draws a hard line between the tactics that get a company its first customers and the systems that later scale sales. Most founders, he says, default to cold email and LinkedIn because those channels are easy to operate from a laptop and because the tooling is abundant. But the first question is not which channel to automate. It is where the buyer actually spends time.

That distinction matters because not every buyer lives in an inbox. A sales leader may be reachable through email and LinkedIn because those are natural work surfaces. A school district administrator, property manager, insurance agent, or truck dispatcher may not work that way. Weak email performance, in those cases, may not mean outbound is hopeless; it may mean the founder is using the wrong surface.

One YC founder selling into a legacy industry spent months on email and LinkedIn with terrible open and reply rates. When he went to an industry trade show and walked the floor, he closed more in three days than he had in three months of cold email. In hindsight, work on subject lines and copy had been the wrong optimization. The customers were on job sites, sometimes at conferences, and reachable by phone.

The practical diagnostic is simple but demanding: describe the buyer’s day concretely. If a founder cannot answer these questions, Kolysh treats that as evidence that they have not spent enough time with real customers.

For the first two or three customers, Max Kolysh says the founder stories he gathered were strikingly consistent: they came from friends in the industry, former colleagues, classmates, or people one introduction away. The reason is not that introductions are a trick. It is that early buyers are taking a bet on the founder as much as on the product.

A new product from a new company requires trust before it has much proof. The order of operations is personal network first, second-degree LinkedIn connections next, and network-search tools after that if needed. One YC founder said she got introductions to about half of the customers she closed during her YC batch through LinkedIn intros.

AI-powered network tools can widen the search. Kolysh points to Happenstance, a YC company that lets founders search across an extended network in natural language. His example query is specific: “find people who work on notification systems at big companies.”

The same specificity applies when asking for the introduction itself. The founder should tell the connector exactly who they want to meet, why that person would care, and what to say in the forwarded email. The easier the request is to execute, the more likely the introduction happens.

The warning is that many founders start setting up outreach tools before they have exhausted obvious network paths. Prospecting tools begin to matter only after a company has 10 to 20 quality customers. Until then, skipping second-degree connections is skipping the lowest-hanging fruit.

A recurring pattern in Max Kolysh’s founder survey is that early customers often closed because founders physically showed up. They did not settle for Zoom. They tried to be in the same room as the buyer, even when that was slower, awkward, expensive, and exposed them to face-to-face rejection.

One founder flew to a single executive buyer four weeks in a row while the buyer kept rescheduling, and eventually closed the account. Another regularly showed up at customer offices uninvited and was asked to leave most of the time. In one case, that founder flew to Hawaii to meet a customer, was kicked out after eight minutes, and through persistence eventually turned that customer into one of his largest accounts.

Small, industry-specific conferences are another high-conversion venue in this account. The suggested playbook is operational rather than abstract: set up a Calendly with 15-minute slots back-to-back for every conference day; email the attendee list before the event to fill as many slots as possible; email again during the event to reach people who missed the first message; and stack meetings throughout the day.

Micro-events can work for the same reason. Some founders ran dinners or happy hours for six to 10 people in their ideal customer profile, at roughly $50 to $100 per head. Founders consistently reported that these converted better than large events. After someone has had dinner with a founder, they are much less likely to ignore a follow-up email, and may even help.

For consumer products and some small-business products, Max Kolysh advises founders to find the online places where potential customers are already complaining about the problem. The point is not to post polished marketing copy into a community. It is to show up as a real person where the pain is visible.

Reddit came up repeatedly in the founder stories. One common pattern: a founder posts a product video, commenters tear it apart, and the launch feels like a disaster. But lurkers who never comment quietly sign up. Another founder got his first 10 customers from Reddit by finding old threads where people had complained about the exact problem he was solving and direct-messaging commenters one by one.

A healthcare founder described responding to Reddit and Facebook complaints as essentially her whole job for a couple of months. She posted two to five times a day, got shadowbanned from some subreddits, and still acquired customers.

The principle extends beyond Reddit. For consumer products, the relevant places may be Facebook groups, Discord servers, or YouTube comments. For B2B niches, they may be industry-specific forums or trade-association message boards. Reddit has one additional advantage in Kolysh’s account: threads are indexed by Google and can persist for years, so useful participation may continue paying off long after the initial work.

After the warm network, in-person rooms, and customer communities have been worked, outbound becomes a narrower job: identify companies that match the ideal customer profile, find the right person, find contact information, and reach out.

Max Kolysh describes Apollo as the most common starting point among the founders he spoke with: a lead database with email finding and a basic sequencer, with a free tier generous enough to build a first list. Clay becomes more relevant when a founder needs research and enrichment workflows on top of a prospect list, such as identifying software usage, hiring activity, or recent LinkedIn posts. LinkedIn Premium remains a rich source of fresh professional data; one founder’s highest-converting cold channel was sending connection requests without a message, then a short DM after acceptance, which led to his biggest customer.

Some of the strongest early outreach was not framed as a sales pitch. It was framed as advice, mentorship, product review, or a whiteboard session. The boundary matters: founders should not disguise sales calls as learning if they are not genuinely open to learning.

The examples are concrete. One founder asked CEOs in his space if they would mentor him; a few accepted, and some later became customers. Another spoke with 200 salespeople before building the product, maxing out LinkedIn connections each week around a specific hypothesis. She reported that about 50% of her connection requests were accepted and 20% of those turned into calls. By launch, she had a pipeline of people who had already told her what they needed.

A dev-tools founder offered startups free whiteboarding sessions on their agent architecture and then opened shared Slack channels to help implement it; the architecture happened to require his product. Another founder selling to lawyers offered to pay $100 to $200 per hour for product feedback, roughly a portion of their hourly fee. About 30% accepted, and the conversion rate made the customer-acquisition cost reasonable, especially for markets with high contract values.

The exact copy matters less than founders often think, but Max Kolysh gives several constraints. Keep the email under 75 words, because long emails are ignored and increasingly assumed to be LLM-written. Make the call to action explicit: reply, call, 15-minute demo, or something else. If the ask is vague, prospects assume the founder will consume too much time.

His simplest test for whether an email sounds human is to read it aloud to a friend. Anything that sounds unlike something the founder would say to a real person should be rewritten. He frames this as a one-minute way to remove the AI-sounding lines that damage cold email.

He also highlights outreach that gives the prospect something before asking for time. An API security company can scan a prospect’s public website and show what it found. A mobile onboarding tool can review a prospect’s app and send a few specific suggestions. A compliance startup prepared short audit notes tailored to each prospect’s product, then asked for a meeting to discuss the findings.

That work is not sustainable at scale, and that is the point. For the first 10 customers, 20 minutes of work before asking for 30 minutes of a prospect’s time is a reasonable trade. Kolysh also recommends following up three or four times over a couple of weeks.

Max Kolysh frames early sales as a staged progression. Customers one through three come from the personal network: friends, former colleagues, and people one introduction away. In his founder survey, he says there were basically no counterexamples to that pattern.

Customers four through 10 come from unscalable work: flying out to customers, sending Reddit DMs, hosting small dinners, writing personalized LinkedIn messages, or offering free consulting sessions. This phase is tedious and manual, but he treats that as part of its value. It lets the founder learn before automating.

Customers 10 through 50 are where the playbook begins to shift. By then, the company is more likely to have a refined pitch, case studies, and a clearer sense of what resonates in the value proposition. That is when Apollo, Clay, email sequences, and higher-volume outreach tools start to make sense: the founder finally has a message worth scaling.