Thirty-four founders, and the one thing that predicted who made it

I have mentored thirty-four young founders. I currently work with twelve of them, and I built and ran the WM Incubator — two months, intensive, eight founders.

Thirty-four is not a large sample. It is large enough to notice something, and small enough that I remember every one of them, which is a trade I would make again.

Here is what I expected to matter, going in: the quality of the idea, the size of the market, and how hard they worked.

#The idea barely mattered

Two founders in the incubator had, on paper, the same business — a service marketplace for a local vertical. Same city. Started within three weeks of each other.

One has customers. The other has a much better deck.

I could give you five more pairs like that. Across thirty-four people, I cannot find a relationship between how good the idea sounded in month one and where the company was in month six. Some of the worst-sounding ideas are now the ones making money, because the founder found out what the business actually was by running it.

This is not the fashionable "ideas are worthless, execution is everything" line, which I think is also wrong. Ideas matter enormously — but the idea that matters is the one you have in month five, and you can only get it by having a worse one in month one and building it.

#Hard work mattered less than I would have guessed

The hardest-working founder I have mentored is not in the top ten by results.

He works genuinely brutal hours. He is not lazy, he is not confused, and he is not doing the wrong things exactly. He is doing a very large number of medium-value things extremely well: the site, the brand, the deck, the automations, the onboarding flow for the customers he does not have yet.

Effort is real, but it is a multiplier, and it multiplies whatever you point it at. Point it at the wrong thing for six months and you get an immaculate version of the wrong thing.

#What actually predicted it

Time-to-first-contact-with-a-real-person who could say no.

That is it. That is the variable.

Not "customer obsession" as a value. A number: how many days between deciding to build the thing and putting it in front of someone who had the power to reject it, and did or did not.

The founders who did that in the first two weeks are, almost without exception, the ones with revenue now. The ones who took three months are mostly still building.

I can be specific about the mechanism, because I watched it happen thirty-four times.

A founder who has not spoken to a customer is operating on a model of the customer built entirely inside their own head. Every decision they make — pricing, feature order, positioning, who it is even for — is a guess compounding on a guess. Six weeks of that produces a product that is internally coherent and externally wrong, and the tragedy is that it is well built. The code is fine. The design is good. It is a beautiful answer to a question nobody asked.

The founder who got rejected in week one has something the other does not: a correction. It hurt, it was embarrassing, and it cost them about forty minutes. Every decision after it is anchored to something real.

The gap between those two people compounds weekly. By month four it is not recoverable by working harder, because the harder-working one is still compounding on the guess.

#Why people don't do it

They know. Every founder I have said this to already knew it. It is in every book.

They do not do it because putting the thing in front of a real person converts an unlimited possible future into one specific piece of information, and usually that information is bad.

While you have not asked, the business could be anything. The moment you ask, it is exactly one thing, and that thing is smaller than the one in your head. Building is comfortable, because building feels like progress and cannot tell you that you are wrong. Asking is the only activity in early-stage company-building that can produce a "no", which is precisely why it is the only one that produces information.

I sold Veridian to two schools and seventy students entirely through cold outreach, with no ad spend. The outreach was not the growth strategy. It was the product strategy — every version after the first was shaped by the specific reasons people gave for not buying the one before.

#What I do differently now as a mentor

I used to open with the business. What is it, who is it for, how does it make money.

Now the first session is one question: who have you shown it to?

If the answer is "nobody yet, I want it to be ready" — and it is that answer maybe seventy per cent of the time — we do not discuss the business at all. We spend the session writing outreach messages, and they send five before we finish.

Almost all of them come back to the second session with a different company. Not because I told them anything. Because someone said no, and told them why.

#The uncomfortable part

Twelve of the thirty-four are still active with me. That means twenty-two are not, and I am not going to pretend all of those are success stories that graduated.

Some stopped. Some had the conversation, heard the answer, and decided they did not want it enough — which is a legitimate outcome and a much cheaper one at seventeen than at twenty-seven. A couple I think I failed, in that I gave them a framework when what they needed was someone to tell them the specific thing they were avoiding.

The ones who made it were not the smartest, and they were not the hardest working. They were the ones who found out they were wrong first.