Note: This is a condensed version of a talk and essay by Paul Graham. I’ve trimmed it down to bare minimum content and added headings for readability. There is paraphrasing/rephrasing based on my own interpretation. Some portions remain verbatum from the original. Please see the source for the full original transcript.
Table of Contents:
Startups are counterintuitive. Knowledge about them hasn’t permeated our culture. You can’t always trust your instincts.
It’s the same kind of counterintuitiveness as in skiing. In order to learn to ski you need to learn new habits.
This talk/essay is a list of the things to remember if you want to prepare yourself to start a startup.
Startups are so weird that if you trust your instincts, you’ll make a lot of mistakes.
The YC partners warn founders about mistakes they’re about to make, founders often ignore them and regret it.
The thing about counterintuitive ideas is they contradict your intuitions, so of course your first impulse is to disregard them.
If founders’ instincts already gave them the right answers, they wouldn’t need YC.
You only need people to give you advice that surprises you.
Your instincts about people are an exception, you can and should trust those.
If you have a bad feeling about a potential cofounder, employee, investor, or acquirer, trust your gut.
This is a case where being self-indulgent is good. Work with people you really like, and who you’ve known long enough to be sure.
The second counterintuitive point is that it’s not that important to know a lot about startups. The way to succeed in a startup is not to be an expert on startups, but to be an expert on your users and the problem you’re solving for them.
If you don’t know anything about how to raise an angel round, don’t feel bad on that account. That sort of thing you can learn when you need to, and forget after you’ve done it.
It’s not just unnecessary to learn in great detail about the mechanics of startups, but possibly dangerous.
If I met an undergrad who knew all about convertible notes, employee agreements, and class FF stock, I would not be impressed, on the contrary it would set off alarms.
Because another of the characteristic mistakes of young founders is to go through the motions of starting a startup; imitating all the outward forms of a startup while neglecting the one thing that’s actually essential: making something people want.
We saw this happen so often that we made up a name for it: playing house.
The reason young founders go through the motions of starting a startup is because that’s what they’ve been trained to do for their whole lives up to that point. Even in college classes most of the work is as artificial as running laps.
I’m not attacking the educational system for being this way. There will always be a certain amount of fakeness in the work you do when you’re being taught something, and if you measure their performance it’s inevitable that people will exploit the difference to the point where much of what you’re measuring is artifacts of the fakeness.
It’s not surprising that young founders’ first impulse on starting a startup is to try to figure out the tricks for winning at this new game.
Since fundraising appears to be the measure of success for startups, they always want to know what the tricks are for convincing investors, and we tell them: make a startup that’s actually doing well, meaning growing fast, and then simply tell investors so.
Then they want to know what the tricks are for growing fast, and we have to tell them: simply make something people want.
So many of the conversations YC partners have with young founders begin with the founder asking “How do we…” and the partner replying “Just…”
Why do the founders always make things so complicated? The reason, I realized, is that they’re looking for the trick.
So this is the third counterintuitive thing to remember about startups: starting a startup is where gaming the system stops working.
Gaming the system may continue to work if you go to work for a big company. Depending on how broken the company is, you can succeed by sucking up to the right people, giving the impression of productivity, and so on.
But that doesn’t work with startups. There is no boss to trick, only users, and all users care about is whether your product does what they want. Startups are as impersonal as physics. You have to make something people want, and you prosper only to the extent you do.
The dangerous thing is, faking does work to some degree on investors.
If you’re super good at sounding like you know what you’re talking about, you can fool investors for at least one and perhaps even two rounds of funding.
But it’s not in your interest to. The company is ultimately doomed. All you’re doing is wasting your own time riding it down.
So stop looking for the trick. There are tricks in startups, as there are in any domain, but they are an order of magnitude less important than solving the real problem.
A founder who knows nothing about fundraising but has made something users love will have an easier time raising money than one who knows every trick in the book but has a flat usage graph.
And more importantly, the founder who has made something users love is the one who will go on to succeed after raising the money.
I think it’s exciting that gaming the system stops working when you start a startup. It’s exciting that there even exist parts of the world where you win by doing good work.
Imagine how depressing the world would be if it were all like school and big companies, where you either have to spend a lot of time on bullshit things or lose to people who do.
But there are parts of the real world where gaming the system matters less than others, and a few where it hardly matters at all, and this variation is one of the most important things to consider when you’re thinking about your future.
How do you win in each type of work, and what would you like to win by doing?
That brings us to our fourth counterintuitive point: startups are all-consuming.
If you start a startup, it will take over your life to a degree you cannot imagine.
And if your startup succeeds, it will take over your life for a long time: for several years at the very least, maybe for a decade, maybe for the rest of your working life. So there is a real opportunity cost here.
Larry Page may seem to have an enviable life, but there are aspects of it that are unenviable.
Basically at 25 he started running as fast as he could and it must seem to him that he hasn’t stopped to catch his breath since.
Every day new shit happens in the Google empire that only the CEO can deal with, and he, as CEO, has to deal with it.
If he goes on vacation for even a week, a whole week’s backlog of shit accumulates.
And he has to bear this uncomplainingly, partly because as the company’s daddy he can never show fear or weakness, and partly because billionaires get less than zero sympathy if they talk about having difficult lives.
Which has the strange side effect that the difficulty of being a successful startup founder is concealed from almost everyone except those who’ve done it.
Y Combinator has now funded several companies that can be called big successes, and in every single case the founders say the same thing.
It never gets any easier. The nature of the problems change.
But the total volume of worry never decreases; if anything it increases.
Starting a successful startup is similar to having kids in that it’s like a button you push that changes your life irrevocably.
While it’s truly wonderful having kids, there are a lot of things that are easier to do before you have them than after, many of which will make you a better parent when you do have kids.
Since you can delay pushing the button for a while, most people in rich countries do.
Yet when it comes to startups, a lot of people seem to think they’re supposed to start them while they’re still in college.
Are you crazy? And what are the universities thinking?
They go out of their way to ensure their students are well supplied with contraceptives, and yet they’re setting up entrepreneurship programs and startup incubators left and right.
To be fair, the universities have their hand forced here. A lot of incoming students are interested in startups.
Universities are, at least de facto, expected to prepare them for their careers.
So students who want to start startups hope universities can teach them about startups, and whether universities can do this or not, there’s some pressure to claim they can, lest they lose applicants to other universities that do.
Can universities teach students about startups? Yes and no.
They can teach students about startups, but as I explained before, this is not what you need to know.
What you need to learn about are the needs of your own users, and you can’t do that until you actually start the company.
So starting a startup is intrinsically something you can only really learn by doing it. And it’s impossible to do that in college, for the reason I just explained: startups take over your life.
You can’t start a startup for real as a student, because if you start a startup for real you’re not a student anymore. You may be nominally a student for a bit, but you won’t even be that for long.
Given this dichotomy, which of the two paths should you take? Be a real student and not start a startup, or start a real startup and not be a student? I can answer that one for you.
Do not start a startup in college.
How to start a startup is just a subset of a bigger problem you’re trying to solve: how to have a good life.
And though starting a startup can be part of a good life for a lot of ambitious people, age 20 is not the optimal time to do it.
Starting a startup is like a brutally fast depth-first search. Most people should still be searching breadth-first at 20.
You can do things in your early 20s that you can’t do as well before or after, like plunge deeply into projects on a whim and travel super cheaply with no sense of a deadline.
For unambitious people, this sort of thing is the dreaded “failure to launch,” but for the ambitious ones it can be an incomparably valuable sort of exploration.
If you start a startup at 20 and you’re sufficiently successful, you’ll never get to do it.
Mark Zuckerberg will never get to bum around a foreign country.
He can do other things most people can’t, like charter jets to fly him to foreign countries.
But success has taken a lot of the serendipity out of his life. Facebook is running him as much as he’s running Facebook.
And while it can be very cool to be in the grip of a project you consider your life’s work, there are advantages to serendipity too, especially early in life. Among other things it gives you more options to choose your life’s work from.
There’s not even a tradeoff here.
You’re not sacrificing anything if you forgo starting a startup at 20, because you’re more likely to succeed if you wait.
In the unlikely case that you’re 20 and one of your side projects takes off like Facebook did, you’ll face a choice of running with it or not, and it may be reasonable to run with it.
But the usual way startups take off is for the founders to make them take off, and it’s gratuitously stupid to do that at 20.
Should you do it at any age? I realize I’ve made startups sound pretty hard. If I haven’t, let me try again: starting a startup is really hard.
What if it’s too hard? How can you tell if you’re up to this challenge?
The answer is the fifth counterintuitive point: you can’t tell.
Unless you’ve had a very strange life you haven’t done much that was like being a startup founder.
Starting a startup will change you a lot.
So what you’re trying to estimate is not just what you are, but what you could grow into, and who can do that?
For the past 9 years it was my job to predict whether people would have what it took to start successful startups.
It was easy to tell how smart they were, and most people reading this will be over that threshold.
The hard part was predicting how tough and ambitious they would become.
There may be no one who has more experience at trying to predict that [than me], so I can tell you how much an expert can know about it: not much.
I learned to keep a completely open mind about which of the startups in each batch would turn out to be the stars.
The founders sometimes think they know.
Some arrive feeling sure they will ace Y Combinator as they’ve aced every one of the (few, artificial, easy) tests they’ve faced in life so far.
Others arrive wondering how they got in and hoping YC doesn’t discover whatever mistake caused it to accept them.
But there is little correlation between founders’ initial attitudes and how well their companies do.
I’ve read that the same is true in the military—that the swaggering recruits are no more likely to turn out to be really tough than the quiet ones.
And probably for the same reason: that the tests involved are so different from the ones in their previous lives.
If you’re absolutely terrified of starting a startup, you probably shouldn’t do it.
But if you’re merely unsure whether you’re up to it, the only way to find out is to try. Just not in college.
So if you want to start a startup one day, what should you do in college? There are only two things you need initially: an idea and cofounders. And the m.o. for getting both is the same.
Which leads to our sixth and last counterintuitive point: the way to get startup ideas is not to try to think of startup ideas.
I’ve written a whole essay on this, so I won’t repeat it all here.
The short version is that if you make a conscious effort to think of startup ideas, the ideas you come up with will not merely be bad, but bad and plausible-sounding, meaning you’ll waste a lot of time on them before realizing they’re bad.
The way to come up with good startup ideas is to take a step back.
Instead of making a conscious effort to think of startup ideas, turn your mind into the type that startup ideas form in without any conscious effort – so unconsciously that you don’t even realize at first that they’re startup ideas.
This is not only possible, it’s how Apple, Yahoo, Google, and Facebook all got started.
None of these companies were even meant to be companies at first. They were all just side projects.
The best startups almost have to start as side projects, because great ideas tend to be such outliers that your conscious mind would reject them as ideas for companies.
How do you turn your mind into the type that startup ideas form in unconsciously? (1) Learn a lot about things that matter, then (2) work on problems that interest you (3) with people you like and respect.
The third part, incidentally, is how you get cofounders at the same time as the idea.
What was special about Brian Chesky and Joe Gebbia was not that they were experts in technology. They were good at design, and perhaps even more importantly, they were good at organizing groups and making projects happen.
So you don’t have to work on technology per se, so long as you work on problems demanding enough to stretch you.
What kind of problems are demanding enough to stretch you? That is very hard to answer in the general case.
History is full of examples of young people who were working on important problems that no one else at the time thought were important, and in particular that their parents didn’t think were important.
On the other hand, history is even fuller of examples of parents who thought their kids were wasting their time and who were right.
So how do you know when you’re working on real stuff? I know how I know.
Real problems are interesting, and I am self-indulgent in the sense that I always want to work on interesting things, even if no one else cares about them (in fact, especially if no one else cares about them), and find it very hard to make myself work on boring things, even if they’re supposed to be important.
My life is full of case after case where I worked on something just because it seemed interesting, and it turned out later to be useful in some worldly way.
Y Combinator itself was something I only did because it seemed interesting. So I seem to have some sort of internal compass that helps me out.
Maybe if I think more about this I can come up with heuristics for recognizing genuinely interesting problems, but for the moment the best I can offer is the hopelessly question-begging advice that if you have a taste for genuinely interesting problems, indulging it energetically is the best way to prepare yourself for a startup. And indeed, probably also the best way to live.
Although I can’t explain in the general case what counts as an interesting problem, I can tell you about a large subset of them.
If you think of technology as something that’s spreading like a sort of fractal stain, every moving point on the edge represents an interesting problem.
So one guaranteed way to turn your mind into the type that has good startup ideas is to get yourself to the leading edge of some technology—to cause yourself, as Paul Buchheit put it, to “live in the future.”
When you reach that point, ideas that will seem to other people uncannily prescient will seem obvious to you. You may not realize they’re startup ideas, but you’ll know they’re something that ought to exist.
For example, back at Harvard in the mid 90s a fellow grad student of my friends Robert and Trevor wrote his own voice over IP software.
He didn’t mean it to be a startup, and he never tried to turn it into one. He just wanted to talk to his girlfriend in Taiwan without paying for long distance calls, and since he was an expert on networks it seemed obvious to him that the way to do it was turn the sound into packets and ship it over the Internet.
He never did any more with his software than talk to his girlfriend, but this is exactly the way the best startups get started.
Strangely enough the optimal thing to do in college if you want to be a successful startup founder is not some sort of new, vocational version of college focused on “entrepreneurship.”
It’s the classic version of college as education for its own sake. If you want to start a startup after college, what you should do in college is learn powerful things.
And if you have genuine intellectual curiosity, that’s what you’ll naturally tend to do if you just follow your own inclinations.
The way to become Larry Page was to become an expert on search.
And the way to become an expert on search was to be driven by genuine curiosity, not some ulterior motive.
At its best, starting a startup is merely an ulterior motive for curiosity.
And you’ll do it best if you introduce the ulterior motive toward the end of the process.
So here is the ultimate advice for young would-be startup founders, boiled down to two words: just learn.
A: If the startup is, if the startup is working in some domain, if it’s not a pure technology startup but is working in some very specific domain, like if it is Uber and the non technical founder was an expert in the limo business then actually then the non technical founder would be doing most of the work. Recruiting drivers and doing whatever else Uber has to do and the technical founder would be just writing the iPhone app which probably less, well iPhone and android app, which is less than half of it. If it’s purely a technical start up the non technical founder does sales and brings coffee and cheeseburgers to the programmer.
A: Basically no, it sounds undiplomatic, but business school was designed to teach people management. Management is a problem that you only have in a startup if you are sufficiently successful. So really what you need to know early on to make a start up successful is developing products. You would be better off going to design school if you would want to go to some sort of school. Although frankly the way to learn how to do it is just to do it. One of the things I got wrong early on is that I advised people who were interested in starting a startup to go work for some other company for a few years before starting their own. Honestly the best way to learn on how to start a startup is just to just try to start it.
You may not be successful but you will learn faster if you just do it. Business schools are trying really hard to do this. They were designed to train the officer core of large companies, which is what business seemed to be back when it was a choice to be either the officer core of large companies or Joe’s Shoe Store. Then there was this new thing, Apple, that started as small as Joe’s Shoe Store and turns into this giant mega company but they were not designed for that world they are good at what they’re good at. They should just do that and screw this whole entrepreneurship thing.
A: Ideally you are successful before you even hire two or three people. Ideally you don’t even have two or three people for quite awhile. When you do the first hires in a startup they are almost like founders. They should be motivated by the same things, they can’t be people you have to manage. This is not like the office, these have to be your peers, you shouldn’t have to manage them much.
A: In the case were you are doing something were you need some super advanced technical thing and there is some boffin that knows this thing and no one else in this world including on how to wipe his mouth. It may be to your advantage to hire said boffin and wipe his mouth for him. As a general rule you want people who are self motivated early on they should just be like founders.
A: I’ll give you two answers to this question. One, ask me questions that are useful to this audience because these people are here to learn how to start startups, and I have more data in my head than anybody else and you’re asking me questions a reporter does because they cannot think of anything interesting to ask. I will answer your question. There is a difference between prices merely being high and a bubble. A bubble is a very specific form of prices being high where people knowingly pay high prices for something in the hope that they will be able to unload it later on some greater fool. That’s what happened in the late 90’s, when VC’s knowingly invested in bullshit startups thinking that they would be able to take those things public and unload them on other retail investors before everything blew up
I was there for that at the epicenter of it all. That is not what is happening today. Prices are high, valuations are high, but valuations being high does not mean a bubble. Every commodity has prices that goes up and down in some sort of cycle. Definitely prices are high. We tell people if you raise money, don’t think the next time you raise money it’s going to be so easy, who knows maybe between now and then the Chinese economy will have exploded then there’s a giant disaster recession. Assume the worst. Bubble or no.
A: Do you mean like IDEO?
Q: No, like Idealab, Garrett Camp’s new one…
A: Oh yeah. There’s this new thing were people start labs that are supposed to spin off startups. It might work, that’s how Twitter started. In fact, I meant Idealab, not IDEO, that was another Freudian slip. Twitter was not Twitter at first. Twitter was a side project at a company called Odeo that was supposed to be in the podcasting business, and you like podcasting business, do those words even grammatically go together? The answer turned out to be no as Evan discovered. As a side project they spun off Twitter and boy was that a dog wagging tail, people are starting these things that are supposed to spin off startups, will it work? Quite possibly if the right people do it. You can’t do it though, because you have to do it with your own money.
A: It probably is true that women have a harder time raising money. I have noticed this empirically and Jessica is just about to publish a bunch of interviews on female founders and a lot of them said that they thought they had a harder time raising money, too. Remember I said the way to raise money? Make your start up actually do well and that’s just especially true in any case if you miss the ideal target from the VC’s point of view in any respect. The way to solve that problem is make the startup do really well. In fact, there was a point a year or two ago when I tweeted this growth graph of this company and I didn’t say who they were. I knew it would get people to start asking and it was actually a female founded startup that was having trouble raising money, but their growth graph was stupendous. So I tweeted it, knowing all these VC’s would start asking me, “Who is that?” Growth graphs have no gender, so if they see the growth graph first, let them fall in love with that. Do well, which is generally good advice for all startups.
A: Literary theory, no just kidding. Honestly, I think I might try and study physics that’s the thing I feel I missed. For some reason, when I was a kid computers were the thing, maybe they still are. I got very excited learning to write code and you can write real programs in your bedroom. You can’t build real accelerators, well maybe you can. Maybe physics, I noticed I sort of look longingly at physics so maybe. I don’t know if that’s going to be helpful starting a startup and I just told you to follow your own curiosity so who cares if it’s helpful, it’ll turn out to be helpful.
A: Having kids is a good way to be efficient. Because you have no time left so if you want to get anything done, the amount of done you do per time is high. Actually many parents, start up founders who have kids have made that point explicitly. They cause you to focus because you have no choice.
I wouldn’t actually recommend having kids just to make you more focused. You know, I don’t think I am very efficient, I have two ways of getting work done. One is during Y Combinator, the way I worked on Y Combinator is I was forced to. I had to set the application deadline, and then people would apply, and then there were all these applications that I had to respond to by a certain time. So I had to read them and I knew if I read them badly, we would get bad startups so I tried really hard to read them well. So I set up this situation that forced me to work. The other kind of work I do is writing essays. And I do that voluntarily, I am walking down the street and the essay starts writing itself in my head. I either force myself to work on less exciting things; I can’t help working on exciting things. I don’t have any useful techniques for making myself efficient. If you work on things you like, you don’t have to force yourself to be efficient.
A: You will know, right. So the question is when you turn a side project into a startup, you will know that it is becoming a real startup when it takes over a alarming large percentage of your life, right. My god I’ve just spent all day working on this thing that’s supposed to be a side project, I am going to fail all of my classes what am I going to do, right. Then maybe it’s turning into a startup.
A: When a start up is growing but not much. Didn’t you tell them they were supposed to read Do Things that Don’t Scale? You sir have not done the readings, you are busted. Because there are four, I wrote a whole essay answered that question and that is to do things that don’t scale. Just go read that, because I can’t remember everything I said. It’s about exactly that problem.
A: Definitely any that will fail. Or if you’ll succeed but your intolerable person. That also Sam would probably sooner do without. Short of that, I cannot think of any, because a large percentage, founders are often surprised by how large a percentage of the problems that start ups have are the same regardless of what type of thing they’re working on. And those tend to be kind of problems that YC helps the most not the ones that are domain specific. Can you think of the class of startups? That YC wouldn’t work for? We had fission and fusion startups in the last batch.
A: If you think of technology as something that’s spreading as a sort of fractal stain. Anything on the edge represents an interesting idea, sounds familiar. Like I said that was the problem, you have correctly identified the thing I didn’t really answer the question were I gave this question begging answer. I said I’m interested in interesting things and you said you were interested in interesting things, work on them and things will work out.
How do you tell what is a real problem? I don’t know, that’s like important enough to write a whole essay about. I don’t know the answer and I probably should write something about that, but I don’t know. I figured out a technique for detecting whether you have a taste for generally interesting problems. Which is whether you find working on boring things intolerable and there are known boring things. Like literary theory and working in middle management in some large company. So if you can tolerate those things, then you must have stupendous self-discipline or you don’t have a taste for genially interesting problems and vice versa.
A: Snapchat? What do I know about Snapchat? We didn’t fund them. I want another question.
A: Starting a startup is where many things will be going wrong. You can’t expect it to be perfect. The advantage is of hiring people you know and like are far greater than the small disadvantage of having some monoculture. You look at it empirically, at all the most successful startups, someone just hires all their pals out of college.
Published October 11, 2014