"Why didn't you go to Stanford?"
It was around 2019, in a final interview at a Silicon Valley venture capital firm. I was doing my MBA at Hult in San Francisco, where I had started the school's venture capital and private equity club and was running it.
I gave them the true answer. I wanted to be done with school in a year. I had not come for the classroom. I had come for the network.
What I was thinking was different. Then why am I in your final round?
By a final interview a firm has seen what there is to see. The résumé, the work, the references. They had seen all of it, and the question left on the table was the name of a school.
They told me to go and work at another fund or an accelerator in the Valley for a while, and then we would talk again.
I was more baffled than anything. But it stayed with me. I went to work in the industry, and in 2022 I co-founded a firm of my own.
Now I sit on the other side of that table, and I hear the question differently. It was telling me something about the person asking.
The number behind the question
Crunchbase tracked just over 4,400 funding rounds of $500,000 or more that went to founders with US university ties in the past year. Startups with Stanford, Harvard or MIT alumni among the founders took more than 30% of them.
3 schools, more than 30% of the rounds. Widen it to the top 7 schools and it is roughly half.
PitchBook counts it another way. In its 2026 ranking, covering January 2015 to August 2026, 1,997 founders who did their undergraduate degree at Stanford raised $252.7 billion across 1,787 companies.
Those founders are good. That is not the point.
The question is what the investor is doing when the school is the first screen.
"We just pattern match"
Ask people in this industry about it and you tend to get the same sentence back. I am not quoting the firm that interviewed me. I am quoting all of us.
"We're not biased. We just pattern match."
The criticism of that line is old, so I will skip to what it is in practice. A committee of strangers assessed this person once, at 18, on grades and essays and a recommendation letter. 15 years later the investor reuses that committee's verdict to decide whether to take a meeting. That is outsourcing.
And it survives the criticism for a simple reason. It is hard to get blamed for backing the Stanford founder who failed. It is easy to get blamed for the unknown one. I do not know what the filter does for returns, and I am not going to pretend I do. I do know what it does for the person using it. LPs do the same thing to fund managers. I went through what that box on the application actually measures in Most VC Track Records Don't Measure What You're Buying.
I used to think the stamp was only a way to cut a long list. My own interview says otherwise. It came out at the end, when the list was already down to me. The stamp is also what people reach for when they have all the information and still are not sure.
So the Stanford question is a statement by the investor. It says: I do not know how to read you directly, so I am borrowing someone else's read, and if I am wrong it will be hard to blame me.
Every market has its own Stanford
Here is what moving between countries taught me.
In Seoul it is a handful of universities. In Tokyo it is the same idea with different names. I now help select founders for programs run with governments, development funds and corporates in Korea, Georgia, Cambodia and Saudi Arabia among others, and each market has had its own version of the question I was asked.
The words change. The function does not.
And here is what gives it away. Most of them stop at the border.
The name that opens doors in Seoul gets a blank look in San Francisco. A founder who is automatically credible at home can land in the next market as nobody, with the same brain, the same product, and the same numbers.
A few stamps do travel. Stanford, Oxbridge, the IITs. That is exactly why Stanford was the one I got asked about. But a global stamp is still a stamp. It is the same shortcut with a longer reach.
Some of what makes a founder good really is local. Language, relationships, knowing how a regulator thinks. Those do not move.
But the product moved. The numbers moved. The judgment that built them moved. And the credibility still reset to zero at the border.
Inside its own market the stamp does carry information. Doors open and calls get returned.
It just tells you more about the room than about the person. A filter that changes at every border is measuring the rooms. Round names do the same thing. I wrote about that in Why Some Startups Raise "Seed" Twice, where one word means a different stage in Korea and in the US.
What the filter costs you
I use shortcuts too. Everyone has a bias, and mine is probably toward founders who sound like me. Instinct alone is just a filter you cannot audit.
Know what you bought, though. When the school is the screen, you are selecting for who was already selected, and you are paying the price that everyone else screening the same way has bid it up to.
The founders outside that screen are unpriced.
The dropout story is the same mistake
The opposite myth is just as lazy. Gates dropped out. Zuckerberg dropped out. So school must not matter.
Zuckerberg left Harvard after Facebook was already live and growing. Gates left with a company to start. Neither of them quit and then went looking for something to do. The work came first, and school got crowded out.
A dropout is a stamp too. Whether you worship the degree or the missing degree, you are reading a label instead of what happened.
What I look at instead
I do not have a clean replacement. But I say the same few things to founders every week.
Has anyone used it. That is the first question, and most weeks the last one. Until other people are using what you built, you do not understand the market, and neither do I. "We have done it", with the hurdles along the way, is worth more to me than the idea.
Then I listen to the words. A founder tells me they are building a learning platform. The ones who have gone deep never say platform. They say they are a MOOC, or a cohort course, or a tutoring marketplace, and they can tell you why it is not the other two. A few sentences in, you can hear whether someone has lived inside the problem or is describing it from the surface.
Then I ask what the customer actually wants, and how far down the founder has gone to find out. It is an onion. The first answer is always the skin. The only version of that question that works: would you buy this yourself? If the honest answer is no, you are building for your own satisfaction, and the customer can tell.
And in a field I do not know, I look for a co-founder who has worked in that field. Not in AI. In the field. A team that is all technology and no industry is three people who have read about the problem.
None of this can be looked up in a minute. The school can. One tells me about a committee in the past. The other tells me about the person in front of me.
Telling those two apart is the job.
If you have ever been screened out by one line that had nothing to do with what you could do, you already know which line it was. Most of the people who wrote that line never found out.
In this newsletter I write down the things I got wrong, or was on the wrong end of, in how capital picks people across borders.