MICHAEL SEIBEL: Shōgun is so interesting because there are all these situations that confront the two main characters where they’re given two bad options. It’s like, “Here are your two options. There are two items on the menu, and they both kind of round to death.”
REID HOFFMAN: That’s Michael Seibel, managing director of Y Combinator. You know, the storied startup accelerator that counts Airbnb, Dropbox, Reddit, and Stripe among its many successful alumni companies. He’s talking about one of his – and my – favorite novels. The classic samurai adventure Shōgun.
SEIBEL: And the smarter of the two characters, Toranaga, he always finds a way to introduce the third option, which is like, “Let’s wait. Let’s wait. How about we wait a day?”
HOFFMAN: Even in the most pressing of circumstances Toranaga takes the time to ask himself this question.
SEIBEL: There was some order given for Toranaga to show up and see the emperor, and then he’s going to be killed. And his brother is dispatched, I believe, to deliver this order, which I love.
That’s a shitty thing, but they’re on this road to see the emperor. It’s going to take 50 days to get there. And so the brother presents him with this order, which in Japan, theoretically, is as good as a death sentence. The brother’s like, “Why do we have to do this long trip? I’ll just kill you right now, and that’s how it should go.”
HOFFMAN: But Toranaga has a different suggestion.
SEIBEL: He was like, “Let’s take the trip.” Fifty days, that’s a long time. Maybe things will change. And so that’s helped me so much along the way when you feel like you’re boxed in, and then you just ask yourself, “Well, do I have to make this decision today?” So often you really don’t.
HOFFMAN: So did asking an unexpected question work out for Toranaga? You’ll have to read Shōgun – or watch the TV series – to find out. But what about you as a founder? How often do you stop and say, “How about we wait?” Especially in a rapidly scaling startup, where speed is of the essence. In fact, even asking this question seems contrary to the classic startup wisdom of moving swiftly and burning your boats. It’s just one of the many contrarian questions Michael urges Y Combinator – or YC – founders to ask themselves.
SEIBEL: Founders always try to reduce multivariable problems into single variables. And a lot of problems are multivariable. Some decisions where you actually do have to decide right now. The other decisions are, absolutely, you don’t. I find the most common question I ask is, “Why are we talking about this? Why is this essential right now at all?” As opposed to, “What’s the right decision to make?”
HOFFMAN: Michael urges founders to ask the uncomfortable, self-searching questions that may feel contrary to the norms of startup life. “Why is this essential right now?” is only one of the many important questions.
Because whether you’re a 17th-century samurai or a 21st-century entrepreneur – knowing what questions to ask, and when, is essential.
That’s why I believe you need to learn to ask the right questions of the right people – and then act upon the answer, even if it seems irrational.
The Third Planet from Altair
Who Killed Harlowe Thrombey?
The Mona Lisa Is Missing!
War with the Mutant Spider Ants
If, like me, you spent part of your childhood enthralled by the book series Choose Your Own Adventure, then some of those titles may be familiar to you.
Every few pages, each of the books – invariably written in the second person – would challenge you to make a choice and then turn to a specific page number to discover the outcome of your decision. Some of these choices were pedestrian: “Do you turn left or right at the fork in the road?” Some had higher stakes: “Do you open the box marked ‘Danger! Mutant Spider Ants,’ or do you douse it in gasoline and set fire to it?”
Even the most unassuming of these questions could have the most astounding consequences for your character – quite often an untimely and gruesome demise.
But in all these books, there’s a way to guarantee success – just flick ahead until you find an ending that appeals to you.
Unfortunately, there’s no such shortcut when it comes to entrepreneurship. You can’t dodge answering questions, and you can’t know in advance how the consequences of your decisions will play out. What you can do is make sure you ask the right questions, of the right people, at the right times.
That’s why I wanted to speak with Michael Seibel. He co-founded the pioneering live video streaming platform Justin.tv, which became Twitch, and also the video sharing app Socialcam. In 2013, he became a partner at Y Combinator – having worked with the accelerator since his Justin.tv days. Famously, it was Michael who introduced Brian Chesky and his then-struggling startup, Airbnb, to Y Combinator.
At key junctures in his life and career, Michael’s decisions have often hinged on asking himself questions that got to the heart of the motivations and assumptions of himself and others.
This ability to ask the “right” counterintuitive questions at the right times of the right people is an essential skill for entrepreneurs, and one that Michael puts at the center of his work with young founders at YC.
Michael first realized the power of asking difficult questions in his high school constitutional law class.
SEIBEL: You’d actually read about the Constitution and read about a lot of U.S. government stuff. And then you would debate resolutions as if you were in Congress. To put it frankly, it was the first time I could measure myself against the smartest kids in my school. And so realizing that I could be as good as them in this debate class, I started asking myself, “Well, why am I not in all the other classes that they’re in?”
HOFFMAN: During these debates, anyone, no matter where they sat in the high-school pecking order, could ask questions of anyone else. And this spirited inquiry led Michael to ask himself a rational question that uncovered the irrational truth at the heart of the school system. It made him doubt the value of the whole high school “tracking” system that grouped kids into classes based upon their perceived ability.
SEIBEL: And up to that point in school, I was in the honors track. That track had just started getting separated into honors and AP. So there just became three tracks, and I was in the middle track, but this class had all three.
And tracking is so weird because on one hand it works great. It puts you in this great environment. On the other hand, if you’re mis-tracked, that doesn’t work so well.
HOFFMAN: Spurred on by this realization, Michael doubled down on his studies and won a place at Yale majoring in political science. In his senior year at Yale, he found himself in a class with a student named Justin Kan.
SEIBEL: And the entire time Justin and I were in school together I thought he was annoying and didn’t want to be friends with him. But we just became friends, and hilariously along the side, Justin and his friend Emmett applied to the first batch of YC.
HOFFMAN: This was the summer of 2005 – and that first batch of YC founders included the Reddit co-founders, and the future Y Combinator president and Masters of Scale guest Sam Altman. The startup that Justin Kan and Emmett Shear brought to YC was a calendar app named Kiko. It was a time when asking yourself, “What should I do with my life?” and answering “entrepreneurship” was a rarity among recent graduates. In fact, doing so was seen as irrational by most people.
SEIBEL: And Justin went down this path of starting a startup, which at the time, I don’t think I can communicate to everyone listening how uncool that was. It was so uncool, it was dismissed. It wasn’t even insulted, because that would be giving it too much attention. It was just irrelevant. Everyone wanted to be a banker and consultant. Sometimes I wish we were still in those days now, but hey, times change.
HOFFMAN: I don’t. I actually think … And you, your whole work, I mean, YC and all the rest. It’s actually better to be builders.
SEIBEL: Fair point, yes. Probably net better for the world, but sometimes I experience the pain of the scene-sters, let’s call them.
HOFFMAN: Yes. Well, the people who are going into entrepreneurship because it’s status versus building.
HOFFMAN: Michael moved to Baltimore to work on the Senate campaign of Kweisi Mfume. Meanwhile, Kiko was crushed by Google Calendar, which launched in 2006. Justin’s next pitch was very different.
SEIBEL: Justin told me they were doing this over dinner with his dad in DC, and we both told him it was the dumbest idea we’ve ever heard in our lives, which I still believe to this day.
HOFFMAN: That idea was Justin.tv. The plan: to livestream Justin’s day-to-day life, 24/7, with a webcam attached to Justin’s head. Today this seems quaint, but in 2006 this was brand-new.
SEIBEL: I think that in hindsight, we always tell founders to start companies to solve personal problems.
And back then he wanted to be an influencer, and there were no influencer platforms.. Twitter didn’t exist. All the things that an influencer would just reach for, he didn’t have. And so almost incredibly, he invented a platform that allowed himself to become an influencer.
HOFFMAN: Note that YC edict that Michael mentioned: YC always tells founders to start companies to solve personal problems. It’s invaluable advice. By asking yourself “Which problem is most personal to me?” you’re far more likely to hone in on a mission that you will have limitless drive to follow. And that’s why, even if that mission seems irrational compared to others you could be following, it really is the one that you should be following.
Justin and Emmett pitched their idea to Paul Graham, serial investor, entrepreneur, and co-founder of Y Combinator. Paul and YC put in $50,000. In the meantime, Kweisi Mfume had lost his bid for the Senate.
SEIBEL: We found ourselves at this interesting moment where I was looking for a vacation, and Justin and Emmett were taking a road trip across the country to start a new company. And I invited myself on this road trip.
HOFFMAN: Michael had begun to ask difficult questions of himself – about his direction, and his purpose.
SEIBEL: I was just going on vacation, and I felt bad because we were in the Bay Area, and they were starting to work. And I’m just the friend who’s sitting there watching their two friends try to do the impossible. So I helped them start a bank account to deposit the investment they had, and I helped them find an apartment to live in, as a thank you. And when I was headed back, Justin gave me a ride to the airport, and he asked me to join the company, and I said no.
And then he was like, “Yeah, just think about it.”
HOFFMAN: If Justin’s question to Michael sounds familiar, cast your mind back to the start of this episode, when we took a trip to Samurai-era Japan:
SEIBEL: Toranaga, he always finds a way to introduce the third option, which is like, “Let’s wait.”
HOFFMAN: It prompted Michael to ask himself another question: “Will taking this path stop me from achieving my goals?”
SEIBEL: The thing that kept on sticking in my mind was: All of my other options would be available in the future. Every other thing I was considering doing would be available, and this is the one thing that felt like it would never happen again, helping my friends start a company. And something just told me, “Do the thing you can never do again when you’re 23 and have nothing going on in your life. Seems like a good play.” And it was. And so I called them, I said, “Yeah, I’m in.”
HOFFMAN: Michael became Justin.tv’s CEO, with Justin as president – thanks to his naive but effective fundraising.
SEIBEL: We figured out how to raise $2 million Series A. And a company that I thought was a bad idea that was not part of my life plan at all, I’m now the CEO of at 23, and we got $2 million in the bank. Fuck, all right. I guess I’m an entrepreneur now. I don’t know.
HOFFMAN: Yes, exactly.
HOFFMAN: After eight months of streaming Justin’s life, the team opened the platform up to let anyone livestream. The company started making a profit. Things were looking good. Until investor Gideon Yu told them his opinion. And he didn’t hold back.
SEIBEL: He was hanging out with us. It was unclear why; he was never going to invest. But he came by the office and he was like, “Oh, you guys look really proud of yourselves. I just want to tell you, your product, as it exists today, is shit. And you and your company are going to die in the next three years — almost no matter what you do — if you keep down this path, and no one will ever remember all the hard work you gave.” And he kind of said it just like that, just kind of like, this is what’s going to happen.
HOFFMAN: It was a brutal assessment, but it got Michael and his co-founders asking themselves the question that they had been avoiding until now. They had been asking, “Is this working now?” to which they could answer, “Yes!” What they should have been asking was, “Will this continue to work, and scale down the road?”
SEIBEL: And it was funny because we knew that, but while the specter of death was over us, because we were on this VC fundraising kind of train, we never really acknowledged it. But the second we were profitable and we got that message, it was like, “Oh yeah, he’s exactly right. He’s exactly right. Let’s fix it.” And in some ways I think Gideon Yu has been my mental model for how to advise startups.
HOFFMAN: Overlooking the most important questions when you’re stuck in the thick of starting a company is a very common mistake. Especially when those questions are a long-term strategy, and you’re caught up with short-term survival. And finding the right people to ask for help when answering those questions will be hard if you haven’t built a strong network. My producers asked me about this while reflecting on my conversation with Michael.
HOFFMAN: It’s very easy to overlook maybe not all of the most important questions, but a most important question. Usually, you’re cognizant of at least some of the really important questions, even when you’re in the thick of getting things running, but that’s part of building the right networks around your company. Because part of the function of these investors, advisors, board members is to go, “Well wait, do you see this risk?” or “Wait, do you see this opportunity?”
HOFFMAN: In this case, Gideon was asking the hard question that Michael and his team hadn’t considered: You’ve put in all this hard work. You’ve secured all this funding. But the product as it exists today isn’t going to get you there. So what are you going to do?
Here’s how Michael and his team answered:
SEIBEL: We decided on two ideas, Instagram for video and Justin TV for gaming. Then we do a classic error, in which we talked to a bunch of investors and said, “Which of these ideas will work?” – which I always love because – you probably should talk to users.
HOFFMAN: Yes, exactly.
HOFFMAN: You might think this classic error stems from not asking yourself: What does my investor know that my customers don’t? This question is important. But not as important as the followup: Does it matter?
Because what your investor thinks you should do may have no bearing on what your customers want. You can think of investors and customers as two different, usually unrelated species. You need both for your company’s ecosystem to thrive; but both operate in very different habitats.
Investors have resources that very few consumers have; they are also idiosyncratic in their outlooks. Consumers tend to be excited about the product or service, whereas investors will typically be more interested in the business model.
Some of your investors will feel they are experts in co-shaping your product. And that’s fine – if they’re right about it. If they aren’t, then you’re going to have serious problems.
Gideon Yu wasn’t an investor in Justin.tv, but his advice prompted Michael and his team to take another look at what their users were energized by.
So in 2011 the Justin.tv team decided to split their focus on two new products. The first was Twitch, a spinoff to capitalize on one cohort of Justin.tv users that had grown rapidly and unexpectedly: gamers. They enjoyed live streaming themselves as they played, and many, many other gamers enjoyed watching. A whole sub-community had grown around video game streaming on Justin.tv.
The second was Socialcam, headed up by Michael. Socialcam was a video-sharing platform aimed at mobile users. Phones had reached the point where mobile video sharing was poised to take off. But for some reason, it was stalling. Michael and his team saw an opportunity.
SEIBEL: We had a good understanding of the problem, which was basically that people felt comfortable taking photos, but not taking video. And the proliferation of smartphones meant that for the first time everyone had a video camera in their pocket. I think that the logical leap we never made was that the way you get people to be more comfortable taking video was not to make it easier to create nice videos. It was to make it very easy to take shitty videos.
And the company that nailed that was Snapchat. Play that hand a hundred times, I don’t know that we make that leap. And I think one of the reasons why was, we weren’t the users. Other things happened, but that was the game and we didn’t win it.
HOFFMAN: And “what is the game” is obviously related to the earlier thing of “ask the right question.” Part of the reason I missed Twitter as an investment is. I went, “Ah, 140 characters, what’s that?” And it was a very similar thing, which is actually, well, no one can look that smart in 140 characters, so no one looks dumb either, right?
So part of how it opens up blogging is, actually writing something where you look like a good essayist, that’s hard, takes a lot of work. Writing something that’s like, “San Dimas High School football rules,” right, that you can write all day long.
SEIBEL: And Reid, we have many peers who prove that point every day on Twitter.
HOFFMAN: Yes, exactly. Exactly.
HOFFMAN: This is why asking the right question is an art. You need to be able to see from multiple viewpoints, understand a range of motivations, dig beneath the surface “how”s to the deeper “why”s. Ask these questions early and often, and then act upon the answers. And if those answers seem irrational, well, that means your competitors are more likely to overlook them.
In 2012, Autodesk made an offer to buy Socialcam. Michael had to ask himself if they could win the game of changing video creation, or if it was time to sell.
SEIBEL: So we knew it wasn’t working, meaning we knew we weren’t changing video creation. And we knew that was the game. I was like, we’re being given a pretty good offer here, I know we don’t have it. We might get it. And we have a lot of investment offers. We could have raised a lot of money and given ourselves a lot of time, but maybe we should get off the train.
HOFFMAN: The question seems like a simple one: should I sell or stay in the game? But there was a deeper question that Michael needed to answer first.
SEIBEL: I had this great conversation with Bing Gordon.
COMPUTER VOICE: Bing Gordon, former chief creative officer at Electronic Arts and VC at Kleiner Perkins.
SEIBEL: He very calmly asked, “Michael, are you a video guy?” And I was like, “What does that mean?”
He’s like, “If we talk 10 years from now, are you going to be working on video?” I literally laughed when he said that. I was like, “Absolutely not. Video’s been the bane of my existence. Video is so expensive. Video has gotten me sued. All the bad stuff in my life has come from video. Absolutely not.” And then it became obvious.
HOFFMAN: So Socialcam sold to Autodesk for $60 million, just 18 months after launching.
SEIBEL: Getting that deal done was a little tricky. It almost didn’t happen, but it did. And we were grateful. It was the best outcome we would have hoped for, for not having the right answer.
HOFFMAN: Socialcam didn’t find the right answer in terms of what their product needed to look like. But Bing Gordon had asked Michael the right question about Michael’s own mission. And it set Michael on course for the next phase of his career: helping founders ask themselves similarly searching questions. In 2014, Michael joined Y Combinator as a full-time partner.
SEIBEL: I had been a part-time partner before and helped out. And, I’d been around YC, basically, my entire entrepreneurial career. So I knew everyone. And then for the first year or so, I really just enjoyed not having a startup that was going to die every second, right? Did a little traveling, I got engaged, I got married, but then one of the things you learn about effective people is they like getting punched in the face. They like work. And so I started doing more and more stuff at YC. And then in 2016, I started running the accelerator, which was just kind of a … I wouldn’t have picked that out, but it’s been going well.
HOFFMAN: A huge part of how Michael works with founders is getting them to ask the right question at the right times.
SEIBEL: So my wife just describes it as therapy, and that’s probably accurate. Nothing that I tell — not nothing, 97% of what I tell a founder, they can find online and read it. I try to be honest. There’s a lot of lies and half-truths about how this game really works.
HOFFMAN: Like therapy, the point of the questions isn’t so much about the answers – it’s about prompting a change in your thinking. I wanted to dig deeper into just how we go about forming the questions we ask ourselves. Brian Christian is the author of a number of books that explore the broader human implications of computer science.
BRIAN CHRISTIAN: One of my favorite quotes about computers comes from Pablo Picasso, who said, “Computers are useless. They can only give you answers.” And I think part of what he is gesturing towards there is something that’s very recognizable to people who work in machine learning today, which is: If you can frame something you’re interested in, in explicitly numerical or computational terms, you can crunch the numbers and get some kind of answer. But it’s much harder to determine what it is that you should be paying attention to in the first place.
HOFFMAN: Our attitudes have come a long way from the 1960s heyday of techno-optimism where computer scientists dreamt of feeding punchcards into a mainframe the size of a hot dog stand to compute every variable in a person’s life, and output the tidy answers to being happier and more productive.
CHRISTIAN: I think there is also a desire to, in some ways, break the tyranny of metrics, that there are certain things that we care about that are not easily encapsulated in some operational, numerical definition. And it’s important to remember that those things are just as real as the things that can be measured, right? That’s the famous quote: “Not everything that counts can be counted, and not everything that can be counted counts.”
HOFFMAN: Asking and answering questions doesn’t just help get us to where we want to be – it actually helps define who we are.
CHRISTIAN: Philosophers like Ruth Chang at Oxford, who says: It is the nature of being human to have to make hard choices based on competing values. That if you have to decide between a career that’s more lucrative, a career that’s more prestigious, and a career that’s more fun, no one can do the math for you and tell you which of those values wins. You have to choose. And she argues moreover that, you know, it’s not like you merely discover within yourself that you are the kind of person who cares more about one of those than the other, but that you become that person by making the choice. And so from that perspective, you know, to struggle with these existential questions of what to do with your life, is an intrinsic part of what it means to be human and, in some ways, a gift to be in that position.
HOFFMAN: At Y Combinator, Michael spends a lot of his time helping entrepreneurs struggling with their own existential crises.
SEIBEL: I try to get them to focus on the right stuff versus the wrong stuff. And a lot of what YC is about is, like, “Here are common patterns of failure that are so well-traveled that we can describe them to you in detail.” We have this line we say, we’re like, “If you’re going to fail, can you just fail in a unique and interesting way that we can learn from?”
HOFFMAN: What do you think are the most classic lies and the half-truths?
SEIBEL: Number one is that money will solve your problems. Number two is that employees will solve your problems. And number three is that smart investors will solve your problems.
Those three things kill YC companies left, right, and center because it gets people to stop worrying about product-market fit and start worrying about company building. And the number one killer of startups is company building. It’s the number one killer of startups. And people are 10x more afraid of scaling too late than scaling too early, which always blows my mind. Scaling too late isn’t fatal. It could be bad, but it’s not fatal. Scaling too early is almost always fatal.
HOFFMAN: You won’t have all the answers, of course. But you do have the questions. Once you figure out what questions are most important to you, you can ask them over and over, in each new situation. The pandemic forced YC to ask difficult questions about its own operation, and if change was needed.
SEIBEL: So I think that the pandemic helped YC a lot. I think that there were so many sacred cows about how YC should run that we just took for granted, because it had been around for over a decade, that had to be killed because we couldn’t meet in person. And it allowed us to innovate, and it’s been great. It’s been absolutely great.
HOFFMAN: Those changes – many of which Michael has spearheaded – included focusing more on virtual engagement, pushing for more diversity, and spreading YC’s international reach.
SEIBEL: I spend more time with the founders going through my group in YC than I did when YC was in person, and my group is two times bigger. And that is not intuitive. We’ve been able to really push forward the secret superpower of YC, which is: fuck the advice, fuck the fundraising, we hope you get customers and make revenue, because at some point, I think — and I don’t know, I’m not a business school guy, but I think — you have to make money, right?
HOFFMAN: Unexpected new circumstances will force you to face questions you’ve been avoiding. Coming up with rapid answers to them – and then implementing these new, possibly uncomfortable, ideas – can help catapult your project ahead.
But there is one ultimate question that every founder faces – particularly when they first start out. It can rear its head at any time, especially during the darkest moments. And it’s an existentially uncomfortable question: Should I be a founder?
HOFFMAN: Who should, and who shouldn’t be founders? What’s a good way for folks who are out there to say, “Oh, I should really consider this and think about this.” And other folks going, “Oh, I should stay away from this.”
SEIBEL: This is a question I’ve really thought about a lot. I think you need to have some type of irrational motivation. And I think the thing that I’ve learned is that there are different sources of irrational motivation. So one can be the irrational motivation of working with a friend. Right? “I really like working with this person. I love working with my best friend. And so I will do the dumbest thing with them, because it’s more fun than anything else.”
Sometimes it’s an irrational motivation about a problem that you have, or your friend or your community has, and solving it. Sometimes it’s an irrational motivation where you’re running away from something. “I can’t function inside of Google,” let’s say. Sometimes it’s this thing broken in your head. The way I describe it is that, “I can only produce my best work when there’s a gun to my head, and I want to produce good work.”
And so I think that I always search for what is the irrational tie, because every founder lies to themselves and tells themself their business is going to work, and can present the logical argument. Right? And we help them write that. And we know that it’s a lie. Right? And so I want to know, when that logical argument starts getting beaten by reality, there’s something underlying that’s going to keep you motivated. And I think that people don’t think about this enough, which is, “What is going to be your — not never-ending but deep — source of personal motivation that will stand bad times?”
HOFFMAN: I guess I give somewhat similar advice, which is: Being a founder means that you are going to make a number of sacrifices in pursuit of this mission. You should not presume that you have the classic 80% of Americans or 85% of Americans think they’re above average, that you have an 85% success rate. You should presume … I usually tell people, it’s like, “Look, you are thinking that you have an over 50% success rate. You should presume a substantially under 50% stance of success rate.” And part of that then is to say, “Look, in your path and approach of doing this, that you’re comfortable, that this was a good use of time if it doesn’t succeed.” Because as you were pointing out, it’s like, “You have some attachment to this.”
And then, even with the right motives, even with the right intention, even with the grit, or the masochism, even with a commitment, you also need to try to get as good a realistic shot as possible on this. And part of that is: Go around to the smartest people you know and ask them what’s wrong with it. Because unless you have something that’s timely, serendipitous, competitive edge, et cetera, these are so much work. Right? Startups always leave lots of blood, sweat, and tears on the floor. These are so much work, that make sure that this is one of the shots that you want to take. It doesn’t have to be the one that’s… They’re not guaranteed, but you want to take this shot.
SEIBEL: Be irrational, and be smart about it, hey?
HOFFMAN: Yes, exactly. Yes. Irrationally good decisions.
HOFFMAN: The way to keep making those irrationally good decisions? Keep asking questions. Not in the spirit of self doubt, but in the spirit of self examination; and in a way that feeds your motivation, and expands your ambition.
I’m Reid Hoffman. Thank you for listening.