Do Not Start a Startup

A friend from college called me last week. He’d been trading at a prop shop for seven years, was having his best year yet, and wanted to talk about leaving to start something. He had an idea — a reasonable one, not a great one — and he wanted to know what I thought. He’d been reading about founders on Twitter. He was tired of staring at screens. His wife was supportive. He wanted to build something of his own.

I told him not to do it.

I don’t think I’ve ever said that to anyone before, at least not out loud. For most of my twenties I was the guy who told everyone to start a company. If you had even the faintest entrepreneurial itch, I’d encourage it. Life is short, the downside is capped, you’ll learn more in two years than in ten at a bank — the usual sermon. I believed all of it. I still believe some of it.

But something has shifted in me over the last year, and I’ve been trying to figure out what.

I think it comes down to the reason. That’s the main thing. I’ve only recently realized how much the reason matters, how much the mission matters. Not because it makes you noble or gives you something to put on a pitch deck, but because it is, in the most literal sense, the only thing that will keep you going.

In every startup there are highs and there are lows, and it is up to you to navigate both. That’s easy to say and almost impossible to understand until you’ve lived through enough of them to see the pattern. The highs are shorter than you think. The lows are longer than you expect. And the thing nobody tells you — or maybe they tell you and you don’t believe it — is that the journey itself is frequently longer than ten years per attempt. Ten years is a long time. Ten years is a marriage, a child growing up, a parent getting old. Ten years is most of the useful career of a person in their thirties.

Over a stretch that long, having fun matters very much. Not fun in the sense of ping-pong tables and offsites. Fun in the sense that on the worst Tuesday of the worst month of the worst quarter, when nothing is working, there is still some part of the problem you actually want to come back to on Wednesday. If that part doesn’t exist, you’re done. It’s just a question of when.

There’s a phase — I think it comes around the fifth or sixth year — when you haven’t really succeeded yet, and you’re still full of self-doubt (or am I the only one?), and the only thing keeping you going is the reason you started in the first place. Every day you feel like giving up. Every day you compare yourself to peers who have surpassed you in ways that are hard to ignore. The college batchmate who’s now a director at a FAANG. The guy you interned with who joined a pre-IPO startup at the right time and just bought a house. The friend at the prop shop who had one good year and cleared more than you’ve paid yourself across the life of the company. You watch their lives from a distance and you do the math on what your effort would have compounded to if you’d spent it somewhere sane.

The only defense you have against this is the reason. If the reason is strong enough, the comparison doesn’t land. If the reason is weak — if you got into it because founders looked cool on Twitter, or because you were tired of your manager, or because your wife was supportive — the comparison will eat you alive. I have seen it eat people alive. I have felt it try.

And the thing about the journey is that it doesn’t stay still while you’re walking it. The market changes. The product you started building is not the product you end up building. Your life situation changes in meaningful ways — you get married, you have a kid, a parent gets sick, a cofounder leaves — and through all of it you still have to grow the fucking thing. The effort remains the same. Only your problems change.

Which is, I’ve come to believe, par for the course. It’s the way this is supposed to be. If you thought you were signing up for a clean problem that would stay clean for a decade, nobody told you what a startup is. A startup is the problem shape-shifting under you every eighteen months while the effort required stays pinned at the maximum.

I should say this out loud, because nobody in my world says it and I think it’s true: starting a startup is not a financially wise decision unless your family already owns a large business.

I mean this literally. If you run the expected value honestly — not the fantasy spreadsheet, the actual one — you are almost always better off grinding at a large company and climbing to the top, joining a pre-IPO startup at the right stage, or, if you have the aptitude for it, sitting at a prop desk and letting good years compound. All of those paths are hard. All of them require luck. But the probability of a meaningful financial outcome is higher on each of them than on founding your own company. This is not controversial among people who have seen the distribution up close. It only sounds controversial because the survivors write the essays.

The reason founding a company can still be worth it is not the money. It’s almost never the money, actually, except for the handful of outliers whose outcomes distort everyone else’s expectations. It’s that you wanted to build this particular thing, with these particular people, badly enough that the alternative — not building it — felt worse than the statistical likelihood of failing. That’s a real reason. It’s maybe the only real reason.

The other thing I’ve come to believe, and this one is more contentious: I think startups should be go big or go home. Running a middling startup on VC capital is almost a kind of crime, and you shouldn’t do it. I don’t mean that morally. I mean it practically. A middling VC-backed startup is the worst of all possible worlds. You have none of the freedom of a bootstrapped business, because your investors own a chunk of your decisions. You have none of the upside of a breakout, because you’re growing at a pace that will never return their fund. You won’t have fun building it, because every board meeting is a quiet negotiation about whether you still deserve to exist. And your investors will never be happy, because the math of venture capital does not reward middling outcomes — it rewards extreme ones, and the middle is where their fund quietly dies.

If you’re going to take venture money, you have to actually be trying to build something enormous. Not “ambitious.” Enormous. If you’re not, take different money, or take no money. The middle is a trap, and I’ve watched good people get caught in it.

So here is the takeaway, the one I gave my friend on the phone: do not start a startup. Unless you love misery and hate yourself. Or have a really, really great idea — one that keeps you up at night not because you’re anxious about it but because you actually want to go work on it at 2am — and a reason for building it that is strong enough to survive five years of self-doubt and peer comparison and a shape-shifting problem.

If you have that, go. Do the thing. I’ll be the loudest person cheering for you.

If you don’t — and most people don’t, and there is no shame in not having it — do something else. Grind at the big company. Join the pre-IPO startup. Stay at the prop desk and get really, really good. Build a life that has room in it for other things. The world has plenty of founders. It has fewer people who are quietly excellent at what they do, going home at a reasonable hour, raising their kids, staying married.

I don’t know if I’m right about this. I might be writing it from inside the fifth-year trough, and in two years I might look back and think I was being melodramatic. That’s possible. I try to hold it honestly.

But when my friend called, I told him what I actually thought, which is that I love this life and I would not recommend it to anyone who asked me to be honest with them. Those two things are not contradictions. They’re the same sentence.