Michael Seibel on Starting a Startup, Finding Product Market Fit, and Fundraising
Michael Seibel is a partner and the CEO of YC. He cofounded Justin.tv, which was in the winter 2007 batch and Socialcam, which was in the winter 2012 batch.
In this episode Michael comments on five of his essays. The essays are: Why Should I Start a Startup?, One Order of Operations for Starting a Startup, The Real Product Market Fit, Users You Don’t Want, and Why Does Your Company Deserve More Money?
0:00 – Intro
2:00 – Three types of people: people highly motivated when working for themselves, people that could succeed starting a startup or within a big company, and people that could succeed within a big company.
6:00 – How do you decide what type of person you are?
7:30 – Identify bias in advice givers
10:30 – Peer advice becomes less valuable during college
15:40 – People aren’t taught how to find ideas
17:20 – Find a particular problem that you’re passionate about
20:55 – Find some friends and brainstorm a solution
23:20 – Build an MVP
25:55 – Two failed orders of operations for starting a startup
30:57 – Why do many founders think they have product market fit when they don’t?
35:42 – Building a successful company is not a single variable problem
37:27 – Socialcam didn’t hit product market fit
38:37 – Justin.tv had $1M in profit before reaching product market fit
42:27 – Some companies take a long time
44:25 – The spectrum of how users are using your product
45:55 – Users that take a lot of customer support time
48:50 – Don’t let the hijack users control the product roadmap
50:31 – A team, a product, and an office are all just a means to an end
51:31 – If you don’t really deserve money, what is an alternate path to create leverage?
53:16 – Breaking even at Justin.tv was a moment of infinite clarity
55:31 – Series A program and leverage
Craig Cannon [00:00] – Hey, how’s it going? This is Craig Cannon and you’re listening to Y Combinator’s Podcast. Today’s episode is with Michael Seibel. Michael’s a partner and the CEO of YC. He co-founded Justin.tv, which was in the Winter 2007 batch and Socialcam which was in the Winter 2012 batch. In this episode, Michael and I go over five of his essays. They’re linked up in the show notes and they are, Why Should I Start A Start-Up? One Order of Operations for Starting A Start-Up, The Real Product/Market Fit, Users You Don’t Want, and Why Does Your Company Deserve More Money? If you have comments, you can reach out to Michael on Twitter @mwseibel. Here we go. Alright, Michael Seibel, so today we’re going to do something different and talk about a few of the essays you’ve worked on in the past, I think these are maybe the past two years?
Michael Seibel [00:51] – Yes!
Craig Cannon [00:52] – The first one is Why Should I Start A Start-Up? You started this essay by saying, “A lot of people ask themselves this question, they often mow over one or more of the following facts, one, the vast majority of start-ups are not successful, two, for talented technical people, it’s relatively easy to get a job and make a large salary, and three, large companies offer opportunities to work on very difficult problems that only often occur at scale. My answer to why you should start a start-up is simple, there’s a certain type of person who only works at their peak capacity when there is no predictable path to follow. The odds of success are low and they have to take personal responsibility for failure. The opposite of most jobs at a large company.”
Michael Seibel [01:37] – Yes.
Craig Cannon [01:38] – Okay, why did you write this essay?
Michael Seibel [01:40] – There are two reasons. One, I talk to a lot of smart technical people, who sometimes feel like they should start a start-up but they’re conflicted ’cause they either don’t know if they have a good idea, don’t have an idea, don’t have a team, and I think they’re really trying to figure out, what should I do with my life? What should my real career path be? When I first started giving people advice, I really assumed that because you could start a start-up, you should start a start-up. If you have the money to afford, if you have the time, if you have the ability, you should be doing it. I quickly realized working at YC that’s a bad idea. What I quickly realized is that there’s probably like three sets of people, there’s probably the set of people who the only thing they can really do is work in a radically entrepreneurial job. Like literally they will not enjoy their life inside of a big company, inside a big bureaucracy. Back in the day, those are the kinds of people that would start small businesses. There’s that type of people and they’re relatively few and far between. I wouldn’t be surprised if that’s only like 1% of the population. The second group of people are the people who are on the fence. They want to be really challenged by work, but they can easily apply themselves in either a entrepreneurial world or in a big company world. Those people are the ones where they have some choices to make. I think a lot of people confuse themselves for this group.
Michael Seibel [03:28] – When in reality, this group is also probably pretty small. Then there’s the last group —
Craig Cannon [03:33] – Pretty small, how small?
Michael Seibel [03:35] – I bet 1%, I bet there’s only —
Craig Cannon [03:37] – Okay so were talking like 2% total so far.
Michael Seibel [03:39] – It’s a really small group who can actually be equally capable in either world. As a founder or as a really effective person inside of a big company. I think you do see this. You do see founders who get acquired, who are actually of really functional inside a big companies and vice versa, right? They go on to start companies and they do well. But few and far between. The last group, and by far the biggest group, are the people who are going to operate best within a large company. The mindset of this type of person at least from what I’ve seen is they want a system to optimize. They want an existing system, an existing path, an existing set of rules they can operate and play by. What’s interesting is that almost everything up to this point in your life has that. Kindergarten through 12th has this. Even college has this and they kind of want to continue that. Someone else defines what’s an A and what’s an F and then I optimize, right? I don’t think there’s any moral judgment to be passed. If most people didn’t feel that way, the world wouldn’t work so it’s a good thing. But there’s this small group of people who feel like I want to make the rules, I don’t want to follow them. When someone’s thinking about whether they should start a start-up, they should really be intellectually honest with themselves about which of these groups they’re in. Anyone who’s trying to guilt trip them into one group or another, it goes both ways. I, in the past would go what to do to be a founder.
Michael Seibel [05:11] – A lot of people’s parents would guilt trip them into being in a big company, or being a doctor, or yada, yada, yada. I just think this is one of these things where you actually have to have some personal, you have to go deep and really understand.
Craig Cannon [05:22] – Did you ever read the book The E Myth?
Michael Seibel [05:24] – No, I haven’t.
Craig Cannon [05:25] – Have you heard of this book?
Michael Seibel [05:26] – No.
Craig Cannon [05:28] – I read it in college, so it’s been a while, but basically the concept was a lot of people who are, you can say craftsmen, so in our context a developer. Often think that they want to become entrepreneurs, so they could do just their craft and just the way they want to, but they don’t realize that as soon as they enter the entrepreneurial realm, most of their jobs no longer matter. This is a common thing. That’s why these indie hackers, people love them so much.
Michael Seibel [05:56] – That’s an interesting point because artists and craftsmen doesn’t equal founder. And even more so, what’s so sad is that as a founder, almost by definition, the thing that you’re good at is the thing that you stop doing.
Craig Cannon [06:09] – Oh yeah!
Michael Seibel [06:09] – And you have to keep on doing the things that you suck at. And be totally okay with it.
Craig Cannon [06:14] – It’s rare that the founder who’s really good at something can delegate so well that the thing they love is still the thing they do. Okay, you get to this at the end. What are the questions, that we’ve had some time since you wrote this one,
Michael Seibel [06:30] – Yep.
Craig Cannon [06:31] – What are the questions that you ask yourself when you have to decide if you’re in that 2% of people?
Michael Seibel [06:39] – I can answer this for myself better. The questions that I asked myself was where am I at my best? And where do I feel like I’m organically applying effort? I’m 100% self-motivated and I think anyone who is 22, 25 has enough life experience to know. And certainly anyone who’s 30 or 30+ definitely has enough life experience to know, in what types of situations am I just naturally, extremely motivated? The other kind of question I asked which is an extension of this is if I think about the moments where I outperformed. Not only was I really extremely motivated, but actually delivered the best results. I did the best. Where does that happen? When I think about those moments for myself, they never really happened in the truly structured things. In the K to 12 setting I did fine in all that stuff. Of course, I did all the classes and got all the grades, but it wasn’t when I was really kickin’ it into high gear. So I think if you’re super honest of when do you really kick it into high gear, that’s when you can start asking this question of yourself, “Who should I be and what career track should I take?” One tricky bit here is you have to be really careful about taking advice from biased people.
Craig Cannon [08:14] – i.e. everyone.
Michael Seibel [08:15] – When I was giving founders this advice, originally I was so biased. I’m at YC, I want you to be a founder. You should be a founder. I was previously giving advice as self-serving. A lot of young people are looking for advice givers and are not doing a good job of identifying bias. I see a lot of people who are generally believing the information that they get from either their peers or from companies that they work at. They don’t realize those people are biased. A perfect example, my brother’s at MIT right now, the most prestigious jobs are jobs at Facebook and Google, and that prestige has actually very little to do with any facts. It actually reminds me of when I was in high school and the crew that I was in, the job everyone wanted to do was be a lawyer. Because we were all government kids. Everyone wanted to be a lawyer. But then if you ask people do you actually want to do the work of a lawyer, you’re like no, of course not.
Craig Cannon [09:21] – No, I just make a bunch of money and I’m a lawyer.
Michael Seibel [09:23] – I want to change the laws and I want to debate the constitution, all these things, but do you actually want to do the lawyer work?
Craig Cannon [09:33] – You’re just driving Model UN right?
Michael Seibel [09:34] – Yeah, exactly! Model UN the the tea right? Similarly when I talked to my brother and I go out at MIT, I meet all these people who have created this imaginary world where when they get to go to Google, they’re going to be on the mission learning team and they’re going to have high-impact from day one. Needless to say, Google loves to pitch this story and people love to eat it. But the reality is when you’re on the other side, that’s often very rarely the case. What I would encourage people to do having been in an offender myself is to just really look for the bias in the person who’s presenting you with options and advice. Understand where their biases are. And understand that Google is operating a machine and the machine has multiple purposes. One purpose is to get the smartest people to work on the hardest tasks, but the other purpose is to get people who are smart and talented so that they’re not working for Google’s competitors and then the last purpose is to get people who are smart and talented so they’re not creating companies that will compete with Google. Everyone understands the first one and they don’t think about two and three.
Craig Cannon [10:46] – No, not at all. Also when I find college kids and I’m totally an offender of this as well, they seek people for advice until the person gives them advice that corroborates with what they want to hear. You just look around and you’re like mmh. Sometimes it feels like you got to do some living. You can only listen to advice for so long.
Michael Seibel [11:10] – The pure advice thing is actually really tricky. It’s really hard to go from, I’d argue that in school, especially like K to 12, a lot of peer advice is very helpful. I learned how to apply to a good school from my peers. My parents hadn’t done it in years. I learned which classes to take, which teachers were good. My peers were like this great intelligence source for so long and it turns out that —
Craig Cannon [11:41] – Yeah?
Michael Seibel [11:43] – With career advice they’re almost horrible. It’s so hard to go from a situation where someone goes from a good source of information to a bad source of information so fast. But that’s what happens in college. Your peers are no longer, because what the reality is is that in many ways there’s a simple track and a simple set of rails all the way up through college. Then college it goes from one track to 1000 tracks and your peers don’t even know anything about any of the 1000 tracks. Suddenly they’re not the experts anymore. Suddenly the kids who’s one year ahead of you really just doesn’t actually know anything more than you do. But your whole old system is still there. Your old system of relying on that is still there. What’s funny is the big companies know this. That’s the cool thing is they are actually relying on you having poor sources of information. Everything that a big company does to attract a college kid is 100% orchestrated.
Craig Cannon [12:50] – There is an information gap there.
Michael Seibel [12:52] – Yes.
Craig Cannon [12:53] – That’s how the business works. You can go and have fun at these jobs. You can go and feel purposeful.
Michael Seibel [13:01] – Of course, what I really get from most people that is the job they should take. But man, if you’re one of those different people, don’t believe the hype.
Craig Cannon [13:10] – This is kind of harsh and I don’t know if it’s 100% accurate but my impression is that the world just tends toward wanting you to be average. You ask someone for advice, they’re like, “These are the known outcomes based on all the inputs you’ve just gave me. You should go work at this company.”
Michael Seibel [13:26] – I wouldn’t say average, I would say wants to put you in a box. It could be a very productive box. It could be a like ridiculously productive box. A well, laid out, well understood, box. They want to plug you in somewhere and move on. That doesn’t work for some people, yeah? Or more importantly, for some people that’s not how you get their best. For some people, it’s almost a guarantee you won’t get their best in that environment. What’s so weird is we’re not talking about do you see a good business opportunity? Or do you have this burning problem? Those things are important but man if you are not irrationally motivated to do start-ups and be entrepreneurial it’s going to be frustrating and you’re not going to understand the risk-reward. Because from a pure money-making perspective it’s probably better to take the job at Google.
Craig Cannon [14:21] – It’s definitely safer.
Michael Seibel [14:22] – Definitely safer, but from a will I enjoy my life perspective, that’s the one for some people, absolutely not.
Craig Cannon [14:30] – Right. But on the other hand I’d just say try it out, see how you feel. You can do plenty of projects in college, you can see how it works, cool. “One Order of Operations for Starting A Start-Up. More often than not when I talk to a technical person who’s thinking about becoming a founder, their number one blocker’s that they don’t have an idea. At some point during their formative years, they learn that every great start-up started with a great idea, and if the idea isn’t amazing, usually as judged by peers, parents, or other people with little stride of experience, the start-up will fail.” Alright, why’d you write this one?
Michael Seibel [15:05] – Think I wrote that.
Craig Cannon [15:06] – I think you did write that one. I think I read that one before.
Michael Seibel [15:09] – I wrote this because this is the number one excuse I hear from people who want to start a start-up and it’s so interesting to me. There are all of these old entrepreneurial rules and business best practices that people just want to apply handing love to a new industry. Tech start-ups might be a little different. Interestingly enough, I don’t even think people are really taught how to find ideas. I feel as though people are sitting there saying, I don’t have an idea and they don’t even really understand the process by which to get one. They think they think that it’s supposed to happen organically, almost like out of the cartoons where the light bulb goes over the top of your head and that’s not really how it works. What I wanted to do with this essay was give people a framework that they could try to use to come up with an idea. Instead of this being a passive process, it can be an active process. One of the things that might be hard for people to see is now being in this world for so long, I have a lot of friends who wanted to do a start-up and didn’t know what they wanted to do and then came up with an idea and were very successful. I’ve seen this happen with a lot of people and so that’s kind of where I’ve come up with this thought.
Craig Cannon [16:33] – You break it out into, one, two, three, four, several steps so I’m going to outline it and then we can explain. “The order of operations, here’s where I’d start. Is there any particular problem that you’re passionate about? The next step, once you decide on a problem, find some friends and brainstorm potential solutions to the problem. The next step, at this point, the most important thing to make that spark turn into a fire is to work together and build and launch and MVP.” That’s basically it.
Michael Seibel [17:05] – It’s not that complex. It’s hard, but it’s not that complex.
Craig Cannon [17:10] – No, it’s not. Let’s start at the beginning. Particular passion.
Michael Seibel [17:18] – I’ll give you the conversation that I had with a founder. I was at a conference over the weekend, called AfroTech and there are a bunch of young people who wanted to start companies there and this woman walked up to me and she said, “I have an idea, I’m doing a start-up that helps other start-ups,” and I call this a meta start-up, and this is start-ups that I kind of don’t like because in some ways, I think it’s like a cop-out. It’s like, hey I want to help give other start-ups advice and guidance but I haven’t really done a start-up myself and I don’t know it’s not great. So I dug in. I was like, so what’s a problem that you have? It was interesting, the woman was a mother, and she said one of the problems I have is that there are a lot of places, when I go somewhere new, there are a lot of places where I don’t know if I can use my stroller with my baby because it’s hard to get around, or if I’m walking on the street, does it have the proper sidewalks, and so and so forth. I’d love like some type of product that would tell me what path I should take that’s stroller friendly. I said to her, “Okay we’re getting somewhere.” Now we’re not in this meta idea, now we’re actually solving a problem directly, great. But I asked her, judge this by two things first. One, how frequent is this problem? And she said, “Well actually, when I’m in an area that I know then I know where to walk, I know how to do it.” So not that frequent. And how intense is this problem? Well, if I have to go over a curb or something, it’s a pain but it’s not intense.
Michael Seibel [19:05] – So then I was like, okay, what if you could solve a problem as addressing one of the top three problems in your life? What would it be? You know what she said? She said, “Affordable childcare.” I was like, you know what? A lot of people have that problem. That’s a big problem! I felt my douche bag investor light go off, like yes, you should do that! That is a problem! Even if you half solve that problem, you’re going to be wildly successful. There are gajillions of people who have that problem. It’s ridiculously intense. It was interesting because in a such a weird way, I feel like she had somehow condensed all of this start-up advice that she’d been given and everything she read to go to this wrong place, when I just asked her what’s the problem in your life? And she almost immediately got to a place that could create real value. Now how should she solve that problem? Okay, yes that’s another question. It’s a personal problem she has and she sure as hell knows that if she makes a product, is it helping her? And at least if it’s helping her, she knows it’s half-way decent. I got to a really exciting place by just digging into what sucks for you? I think everyone can do that exercise. I think that exercise can be in form by your work life or your personal life. Many start-ups I see are created because they were doing something at work and they were like, this sucks! Why do I have to keep on doing this? I’m going to make this thing and then sell it to my job so that no one has to do this ever again. So you can apply either way.
Craig Cannon [20:37] – Right, and this is a very easy way to get people to pay you. Whereas doing things like strapping a webcam to your head, not as obvious of a problem.
Michael Seibel [20:45] – Pretty non-obvious.
Craig Cannon [20:51] – So “once you find a problem, find some friends and brainstorm potential solutions to the problem.”
Michael Seibel [20:55] – Here’s the trick. You don’t need to skip to a solution too early. Your thought process should be whatever solution I come up with is probably going to be wrong. What my solution really is, is a first hypothesis and basically it’s like, imagine I am trying to, I don’t know redo some famous science experiment. Like I’m trying to figure out the speed of gravity or something. I first have to start with a hypothesis and then test and then repeat, repeat, repeat until I get to my answer. I don’t need to know the answer up front. That’s not the goal. If you have a big enough problem trust that if you come up with anything resembling an answer in a reasonable amount of time, years, you have a big business. Instead of thinking, okay let me come up with an idea on how to solve it, I think the more important thing is to start talking to other people about it. Because I think that the number one thing you have to do to create a company or one of the top three is to have some co-founders. It’s really helpful to figure out who the types of people who you can talk to about something and you riff well. You enter each other’s ideas of, there’s some type of people you brainstorm with and I say Craig wouldn’t it be cool if the sky is red and Craig says no. And that’s the end of the brainstorm. Whoa, we might not have the best founder relationship. As opposed to, wouldn’t it be cool if the sky was red? And Craig says, you know we can’t make the sky red but what happens if we tint all the windows in the house red so you can make it look like the sky’s red?
Michael Seibel [22:37] – And you’re like, oh maybe, but we’re moving right? You want to start having those conversations with people that you like. If you’re not technical people who are technical people who you just respect their intelligence. Ideally together you start coming up with an idea. What’s cool about that is that if you come up with an idea together, everyone one that you’ve been talking to feels some ownership of the idea. And that helps kind of light that flame.
Craig Cannon [23:08] – Yeah it’s dangerous to be married to a solution.
Michael Seibel [23:11] – Yes.
Craig Cannon [23:12] – That is a real problem. Alright, so “at this point, the most important thing is to turn that spark into a fire, and work together to build an MVP.”
Michael Seibel [23:21] – Boom! So biggest mistake that people make at this point, is to think that the things they need to do is, now I feel like some weird, phantom MBA starts whispering in their ear and is like, “You need to incorporate, you need to raise the money, you need to da, da, da.” Kick that person out. That’s not what you need to do. Can you just build a simple V1 of the product? Can you build the simplest V1? Even if it’s just an Excel spreadsheet. Something powered by the Excel spreadsheet. Give it a try. The way to get that team even more excited is to try something. Oftentimes I see people get lost in that legal or that fundraising world and never even got to give their thing a try! You don’t need bank accounts to give something a try. You don’t need incorporation to give something a try, for the most part. There’s some ideas you do but for the most part you don’t. A way that you can really get that team solidified is just to get some kind of MVP and get one customer, two customers. This is one, the final test, unlike hey can we work together well? And two, man it amps up the excitement a lot and so that when you have to go through some of the bullshit of incorporation and stuff or fundraising, the excitement level’s high. Also third, it gives you leverage. If you actually have something with a couple of customers then investors or farmer should like to talk to you, and if you just have an idea.
Craig Cannon [24:46] – Especially people who are on the fence about working with you.
Michael Seibel [24:49] – Yes!
Craig Cannon [24:50] – All of these things that you’re saying, these are all excuses you create because it’s easier than working on the actual problem.
Michael Seibel [24:57] – Well, yes for some people, but I also think it’s, for some people it’s what they think is the right path. Some people are being, I don’t want to say lazy, but some people are being I don’t know, some people are kind of being disingenuous in this process but I think other people actually think, “Oh no the right thing to do is actually to come up with an idea, fundraise, outsource an MVP, and then hire a team.” Some people think that’s the right thing to do and it’s like that’s a harder path. You don’t have to do that.
Craig Cannon [25:34] – Well, my failure mode is different then. Then you wrap this essay up with a couple of things. “The two failed orders or operations.” Well, these are two of many I suppose.
Michael Seibel [25:46] – Yes.
Craig Cannon [25:48] – “One, come up with an idea and then pitch to investors. Two, I have an idea but can’t write code and none of my co-founders can either.”
Michael Seibel [26:00] – We’re in a time where there are a lot of start-ups and there’s unfortunately, your people that you’re pitching, if they’re good, they’re getting pitched by a lot of people. You need to bring some real fire to that meeting. A good deck is no longer good enough. I would argue that back in the day, like in the 90’s, it was so expensive to get software made and hosted, and online and get users that a good deck, or actually back then it was like 100 page business plan, that’s all you could get, that’s all you could do with that money. We’re not there now. You’re not bringing the fire if you just bring in an idea. Now some people can raise interest in an idea. If you’re one of those lucky people, that’s great! But I don’t know, I’d rather be good than lucky. In terms of knowing how to talk to a co-founder. This is a constant issue and I’m surprised by this. Look, I talked to a friend the other day who is thinking about starting something and doesn’t have a technical co-founder yet and even she was a little concerned about getting one even though she’s kind of been in the tech start-up world. I had to encourage her. I had to tell her no you can. You can do this. Not only is this not going to be that hard, it’s probably going to be easier than outsourcing. The one very mode that I see a lot, is that I see people who can raise a little bit of money, go down the outsourcing route, and believe that they are progressing. They believe that the thing is moving forward and what they don’t realize is that, for the most part, they’re going to need to iterate
Michael Seibel [27:48] – anything that’s created by those outsourcers over and over again and it gets pretty expensive pretty fast. The thing that I see so constantly is outsourcing iterating with an outsourcer is you run out of money. You haven’t figured things out yet. You have to go to an investor and say well I don’t have anything, I need more money. That’s death. What’s so sad about that death is that you’ve been kind of walking that path the whole time but often you don’t realize it ’til the very end and after you’ve spent money. And sometimes it’s people’s own money. They do their savings, they hand their savings to these people. I don’t think the outsourcing folks are bad or morally wrong, I just think that people should take the hustle that they have and apply it to the number one problem which is it just turns out that getting a technical co-founder tends to be cheaper than paying an outsourcer.
Craig Cannon [28:46] – To push back, if you had someone outsource something and that got you a couple of customers, then you found a co-founder, that’s a great outcome.
Michael Seibel [28:58] – Well, it’s interesting because I think that a lot of people go into the outsourcing route with that intention but the thing that they don’t realize is that they didn’t feel comfortable going and finding a technical co-founder now when they don’t have any traction and they don’t have a product. Most likely, the outcome is the outsourcer will give them a shitty product, a shitty V1, they won’t have any users then, and now they’ll go after technical co-founder with a shitty product and no users and less money.
Craig Cannon [29:30] – They want to give the co-founder 10%.
Michael Seibel [29:32] – They’re not actually in a better position! Yes, I think that one, and I mentioned in the essay, there are counter examples, there are people who do go from contractor to big company but it’s just I would argue it’s a harder path and it’s a path that’s much more likely to lose you a lot of money.
Craig Cannon [29:55] – Your own money.
Michael Seibel [29:55] – Your own money.
Craig Cannon [29:57] – “The Real Product/Market Fit.”
Michael Seibel [29:59] – Yes
Craig Cannon [30:00] – This was a good one. Not that the other ones aren’t great.
Michael Seibel [30:04] – I try my best.
Craig Cannon [30:07] – “I often talk to founders who believe they’ve found product/market fit when they haven’t. This is a huge problem because they start hiring people, increasing burn and optimizing their product before they’ve actually discovered what needs to be built. I’m writing this post to help you understand when you’ve really found product/market fit. To start, read Marc Andreessen’s On Product/Market Fits for Start-Ups. It has been the single most influential post for me as an entrepreneur and was the first time I ever read the term. Here’s how he defines the term: The customers are buying the product just as fast as you can make it or usage is growing just as fast as you can add more servers. Money from customers is piling up in your company checking account. You’re hiring sales and customer support staff as fast as you can.” Why do most people think they’re there when they’re not?
Michael Seibel [30:55] – One, it’s intellectually convenient to. A lot of founders are really excited at the prospect of company building and they think that company building is what creates success, and by company building I mean hiring employees, and having great culture and getting an office and having management and so on so forth and they are not real with themselves that the real challenge is to solve a problem. The company building happens, for the most part, after you figure out how to solve the problem, not before. I hear this term product/market fit so often and I have to tell you 98% of the of the time it’s used incorrectly. What’s so frustrating is that people almost act like it’s an undefined or flexibly defined term and it’s totally not. It’s like saying green, blue, yellow, we can call all of those orange. It’s okay, we’ll just call them orange.
Craig Cannon [31:53] – It’s different for every company.
Michael Seibel [31:54] – Yeah, exactly! It’s like it’s your interpretation of it. No, it’s not! That’s not the case, there’s a defined term. Make up another term if you want to mean something else. The most common way it’s mistaken is, it’s weird because it sounds so close, but it’s not it. It’s, I built the thing that customers want.
Craig Cannon [32:14] – Right.
Michael Seibel [32:15] – What’s hilarious is product/market fit is what happens after you build the things that customers want. It turns out the only way you know you’ve built something the customers want, is because they’re using it in an explosive and destructive way. People want to separate these two concepts. It’s so amazing! You can seem intellectually wise, it’s just so much easier to be able to look at your thing and say, this is what customers want and not have to really have any customers.
Craig Cannon [32:42] – Oh yeah, totally.
Michael Seibel [32:42] – It’s easier to say that. Man, people really just want to separate those two things out and it’s like, if you are not getting explosive usage, you do not have what customers want. Or there aren’t that many customers in which case you don’t have a big business. The awful reality is the vast majority of founders, the vast majority of YC founders even, never find product/market fit ever. I’d argue that more acquisitions then you might know are of companies that did not find product/market fit. Once you find product/market fit, I almost say it’s your company to fuck up. It’s almost like, this is going to work unless you screw it up. Whereas like pre-product/market fit, there’s all this stuff you need to do to even see if there can be something there and you really don’t know. Post product/market fit is like, if you execute, you get there. Pre-product/market fit you can actually do great and never have anything.
Craig Cannon [33:48] – Moreover, you can still have customers, you can still have growth if you’re pre-product/market fit.
Michael Seibel [33:53] – Yes! Yes! It’s not like when I go from zero customers to one customer, I’ve not hit product/market fit. Marc does a good job of defining it because it’s experiential. I think if you tried to define it any other way, people would find loopholes. They’ve clearly already found loopholes, but it’s like is the growth killing you? If I were to extract what the meaning of this sentence, is the growth almost killing you? And is it profitable? People always want to forget the second one. Money is piling up in the checking account —
Craig Cannon [34:26] – That is by definition, profit.
Michael Seibel [34:28] – Literally there are so many companies who are like, the growth is killing us, and I’m like oh, show me, explain and they’re scaling negative margins. They’re saying, hey Craig how about you pay $0.75 and I give you a dollar worth of value?
Craig Cannon [34:44] – That’s amazing.
Michael Seibel [34:45] – Craig would make that trade, 17 times and hour easy.
Craig Cannon [34:49] – I love the PV quote, where they figure it out before you. Because you’re just burning VC money or your own money.
Michael Seibel [35:00] – The customers are known for those types of deals. Because if you have the problem, you really understand the value of the solution. What’s funny to me is that oftentimes founders will want to try to reduce this. I get this question so often it’s like, “Should I optimize for growth? Should I optimize for retention? Or should I optimize for profitability?” And my answer’s always the same, yes. What makes you think that building a successful company is a single variable problem? How could that be the case? It’s clearly not the case.
Craig Cannon [35:44] – Well, especially in the venture paradigm.
Michael Seibel [35:46] – Yes! For a tech start-up, yes, it’s clearly not the case. There’s clearly multiple variables here. One of the things we try to talk to people. Often if you’re charging your customer, we try to talk to you about revenue. The reason why is in many cases, like revenue that you keep is as close as we can get to a single metric but even that is not perfect. You really want to know what percent of your revenue are you losing every month? You really want to know how much does it cost you to get that revenue? You really want to know how satisfied are your customers. You need to know those other stats to really know. Product/market fit, I wish more comfortables, I talk about pre-product/market fit as just being in the suck. It’s just like, you’re in this river of crap and you’re not enjoying your life. Unfortunately, every start-up, I almost imagine, this isn’t the case but I almost imagine that there’s like a pre-determined amount of time you have to live there until you figure out how to get out, and a lot of founders just want to cheat. A lot of founders are just like I want to get out by cheating. I just want to pretend like I’m not in here and build a nice house or I just want to do something that’s basically unprofitable so I can pretend to be out of here. You can’t pretend your way out of product/market fit. It does not work. You have to be comfortable, you have to be so passionate about the people you’re working with, and/or the problem that you’re solving that you are comfortable failing at it for a while.
Craig Cannon [37:22] – Did you guys hit it at Socialcam?
Michael Seibel [37:24] – Absolutely not!
Craig Cannon [37:25] – You’re the perfect example.
Michael Seibel [37:27] – Socialcam over the course of four months got 16 million downloads, got one-eighth of every single person on Facebook to watch at least one video on our platform, over the course of three to four months. Wouldn’t that be like almost the definition of product/market fit? We certainly were dying in the traffic. We shut off half the world couldn’t use the product because we were like, we can’t do it. Absolutely not. For our product, one we had no monetization. We probably could’ve figured out monetization but at that point, no monetization. Two, more importantly we had horrible retention. Absolutely horrible retention! If you came and downloaded the app, the chances that you were using that app 10 days later was basically zero. If you watched a video, the chances that you’d watch five more videos that week, very low. It was interesting because it was like we had an exit. We had tons of growth. We for a while were top five in the app store, yet not product/market fit. I’d even argue, I mean if you look back at Justin.tv. Justin.tv, by let’s say two thousand and let’s think, ten. Starting in 2006, by 2010, let me make sure I have those dates right. I think that’s correct. We made $8,000,000 in revenue, 1,000,000 in profit with approximately 30,000,000 monthly viewers on the product. We were not a product/market fit.
Craig Cannon [39:10] – It’s tricky.
Michael Seibel [39:11] – Yeah! It is hard. The reason why for the Justin.tv case, was that we didn’t have a repeatable engine to create growth. That 300,000,000 monthly uniques, we couldn’t make that number go up. The second thing was that, unfortunately we had tons of copyrighted content on the site. So we weren’t able to reliably monetize that over time. It’s funny because I can’t imagine how many pitches that I gave trying to present that as a good business when in my heart, I knew it wasn’t a good business. It wasn’t until we actually found Twitch and found gamers, did we actually find something that had this repeatable machine that both spit out more users and more money and retention. What was so interesting is that we’d solved all the technology problems at Justin.tv.
Craig Cannon [40:12] – Well, clearly the market does not care what you have behind the hood.
Michael Seibel [40:16] – No! No! We had all the live video streaming, we had all the chat, we had everything!
Craig Cannon [40:21] – No.
Michael Seibel [40:22] – Let’s be clear. Necessary, but not sufficient. We never could’ve done it without that stuff but it was not sufficient at all. What did that take? It took six years for us to have a hint that that’s what the customer base was.
Craig Cannon [40:38] – Because then the suck period with Twitch as game streaming, how long did that go for?
Michael Seibel [40:44] – Not every long.
Craig Cannon [40:45] – Not very long?
Michael Seibel [40:46] – You know why? It was interesting, it didn’t go for very long because everything else was already, the table was set.
Craig Cannon [40:51] – You were kind of in place for the marketing.
Michael Seibel [40:52] – We had people, we had all the technology we needed. We had money ’cause we were running profitably. The table was set and then bam, the food arrived. Everyone was seated, the table was set, bam the food arrived, the meal happened. But we were waiting at the table for like six years. We built the tables, we had the door open, we sent invites out and no one cared.
Craig Cannon [41:19] – But if you started Twitch in 2006, it probably wouldn’t have worked.
Michael Seibel [41:22] – No. I don’t think it would’ve worked in 2006. Looking back, so many little things matter. One of them was webcams. Webcams went from being something you had to buy extra, to something that were built into every laptop. Little things like that can change things. Had we been talking to our users more, we could’ve gotten there sooner. We could’ve gotten there by like 2008, 2009? We didn’t die, have another shot.
Craig Cannon [42:03] – That’s a big takeaway. Which is kind of a separate thing, probably a separate essay but, learning when you should’ve quit, maybe. Some people do hold onto things.
Michael Seibel [42:13] – Some people hold on to things, it’s funny. I see it two ways. Some people hold on to things, on the flip side though, man, there are still some companies some companies take a long time. There are some peers of ours who now are really doing well but just took even longer than we did. If you’re willing to put in the time and grind it out, every year you do not die, your chance of being successful go up.
Craig Cannon [42:47] – Then some companies sell too early and they could’ve been so cool and then they die within big companies.
Michael Seibel [42:53] – That happens too.
Craig Cannon [42:54] – People trying to get paid. “Users You Don’t Want.” This one was contentious.
Michael Seibel [42:59] – Yeah, this one’s fun.
Craig Cannon [43:00] – Let’s bring it back! “When you’re just getting started, many start-ups will take every user they can get. They have a strong idea of a problem and they want to attract as many users with that problem as possible. Unfortunately, when you open up the barn doors you get all sorts of people with all sorts of problems. Some of them will try to hijack your product to solve a problem you didn’t intend to solve. By and large, these hijackers are users you do not want.” When has this happened to you?
Michael Seibel [43:30] – Justin.tv was by definition hijacked. We built a product to allow people to livestream their lives, and within a year it was being used to stream copywritten content around the world. Absolutely the definition of a hijacked product. What’s interesting is that, I’m a little afraid of my phrasing here, I wish I didn’t say that these hijack users are not users you want because it turns out that sometimes they are. The reason why we even had an inkling to do Twitch, was because some percentage of the hijack users, were video gamers and it turns out that that could’ve been a much bigger community if we helped it and it eventually became one. Users using your product for a whole variety of things, I like to think about it more on a spectrum. There’s the user who’s using your product as intended. Great, it makes you feel good. Maybe there’s a business there maybe there isn’t. There’s the user who’s using your product in interesting ways with potential. Study those users. Those users are very important. Video gamers of Justin.tv. There are users who are using your product with ways that it’s extremely clear to you that there isn’t long-term value, even if there is short-term value. I’d argue that those are the copper… streamers on Justin.tv. There are short-term value because we can monetize with ads and so and so, but long-term we weren’t really creating value. And then there are hijack users. I’d argue those are users who are using your product and they’re creating no value and they’re probably actively decreasing your value.
Michael Seibel [45:20] – They’re harming your network. For us, that would be people like, some people stream pretty horrible things on Justin.tv and that was not good. What’s interesting is that now when I talk to start-ups at YC, they don’t know to recognize these hijack users and just kick them out. No one talks about these people. They always talk about everyone before them. Which those are users that have some value. I’m talking about the hijack users. If you can identify the hijack users you need to kick them out. They’re actually going to hurt you more than they’re going to help you. This is important and I think that it’s not that the hijack users have bad intentions even. They can, some do. But some it’s just they want to use your product to do what they want to do, and that is literally going to hurt your business. There are some things that what people do on your platform, it hurts your business.
Craig Cannon [46:22] – How would you classify someone who just takes up all the customer support time?
Michael Seibel [46:30] – It depends on how much revenue they’re generating. I’d argue that if they are generating enough revenue to justify that support time, great! If you have a have a free product?
Craig Cannon [46:40] – Not so much.
Michael Seibel [46:41] – Not so much, probably right? Especially if you’re not learning. If they’re taking up a bunch of your support time, but you’re learning and you’re actually improving your product because of it, great. For example, let’s say, I’ll create a hypothetical example off the top of my head. Imagine that in the beginning of Airbnb, someone wanted to use Airbnb to host drug parties. The illegal drug parties like crack parties. They thought of these Airbnbs as like mobile crack den. Well, let’s think about this for a second. On one hand, well they are paying.
Craig Cannon [47:24] – Yeah, they’re paying for the big apartments!
Michael Seibel [47:26] – They’re paying for the nice places. Yeah, they got money, right? On the other hand, they’re probably destroying those apartments doing something highly illegal, getting the police called on them, like da, da, da. These are hijack users, these are not the types of people you want. You can be a hijack user and there’s still some positive. But if there’s way too much negative, it’s not worth the positive, you know? I actually see it a lot where the hijack user’s paying, so you feel bad. Because when the hijack user is not paying, usually you feel less bad kicking them out. When they’re paying, you’re like, well but they’re paying and it’s like no. You should still kick them out.
Craig Cannon [48:14] – Especially in the context of you have these metrics you want to hit —
Michael Seibel [48:17] – Yes, right, oh my god like this is real stuff but it’s —
Craig Cannon [48:22] – Shut it down. You end it in a nice clean way by saying, “By focusing on solving one problem really well, you’re betting on making a small amount of people very happy. If you let any user that walks in the door of the product roadmap, you’re going to end up doing a shitty job at half-solving a lot of problems.”
Michael Seibel [48:40] – To be clear with this ending, my whole goal is, like I said, there’s like this spectrum of potentially useful users. My whole goal is don’t let the hijack users, steal your product roadmap. It’s really totally okay if you want to explore some of those use cases that are not the use cases that you thought, that are still use cases that could be good for you. It’s totally okay to explore to experiment so and so forth, but don’t let the crack den guys on Airbnb control the product roadmap for Airbnb and suggest the features that they want, it’s not where you want to go.
Craig Cannon [49:19] – This is sort of the mentality, at the end of the day, you have to have some opinion about your product. You can’t just be like, I don’t know, the market’s going to take it wherever, the customer’s take it wherever.
Michael Seibel [49:27] – Yes, it’s really hard to build product without some opinion.
Craig Cannon [49:32] – “Why Does Your Company Deserve More Money?” Dun-dun. “The hardest conversation I have to have with a founder, is when they’ve spent their one to two million dollar angel round but haven’t found product/market fit. Unfortunately, I have to ask them a very unforgiving question, why does your company deserve more money?”
Michael Seibel [49:49] – This is so hard. This is really hard for YC companies. In one way, we try to make YC companies feel very special and extremely the investing community often makes YC companies feel really special. But man, isn’t weird that someone is given $2,000,000 to do something and doesn’t succeed thinks, you know what, I need another $2,000,000. Part of me just wants to bring people back to the real world. Maybe you do. Maybe the next $2,000,000 you’re going to get, make it work! I’m not willing to say it won’t, but I think we should just, it’s certainly a lot easier to raise that money if you’ve done something for the first $2,000,000. Sometimes founders, once again, sometimes founders cargo cult. They think, oh we have this team, we built a team. We have this product —
Craig Cannon [50:44] – We’ve got a cool office.
Michael Seibel [50:45] – We have a cool office, look at what we’ve done! Nobody’s grading you on those things. Those are means to an end, not an end. You can’t be like, “Oh I’m an NFL coach and look we have a team, we have this amazing stadium, you should renew me,”and it’s like, well the record of the team is zero wins. You can’t get another contract if you have no wins.
Craig Cannon [51:07] – Moreover if you, I mean you might have all the MVPs but if they can’t play together —
Michael Seibel [51:11] – Then no wins. People like to confuse means for ends. Means are a lot easier to get. I can go out and get an office. I can go and hire, right? Solving people’s problems, that’s hard. A lot of the time I have to have these conversations with founders where I’m just like, look what is an alternative path? If you don’t really deserve money right now, what is an alternative path? And the sad but true fact is cutting burn and trying to break even is more often than not the right thing to do if you haven’t hit product/market fit for that first one to two million dollars. More often, that’s what’s going to create that leverage. You get to break even and that gives you time to figure things out. Like we said before, not having product/market fit doesn’t mean that you’re not growing, doesn’t mean you’re not generating revenue, it just means that you’re taking off. Oftentimes, sometimes you just need more time. But asking investors for that time is oftentimes way less fruitful than just cutting, burn and giving yourself that time with the revenue that you’re generating.
Craig Cannon [52:21] – Exactly.
Michael Seibel [52:24] – Man, I learned this the hard way with Justin.tv. Sometimes I just think of Justin.tv, I just had myself completely fooled. I was out there pitching a site, that half of the usage was spreading copywritten content, and it was like a public site! Any investor could just, before the meeting, go to our website and in like three clicks see content we didn’t have the rights to! And then I would go in and try to pitch them on how I was going to be a billion dollar business. I don’t even know how I did it, looking back.
Craig Cannon [52:59] – You seem to have made every mistake.
Michael Seibel [53:01] – Many mistakes. I like to write about things that I have personal mistakes and experience. But I will tell you that when we broke even at Justin.tv, that was the moment of infinite clarity. What was weird is that looking back, one of the things I see in other founders and I recognize in myself is that, when you’re operating on the investor’s dime, oftentimes you’re trying to optimize what the investor wants to hear. Oftentimes, you’re trying to structure your pitch and your strategy to what’s going to get us more money? As it turns out, the investor’s really your customer, because they’re the only one giving
Michael Seibel [53:38] – you the money you need to survive. Your users become secondary. When you break even, there’s this magical moment where you realize, wait a second, I can just generate money from my users. I don’t need these investors anymore. And strangely, weirdly the only group of users who, the only group of people who are harder to understand than users are investors. For many people, some people can just spin investors, some people have a way with investors, but many people and investors think they’re this really confusing group and it’s like, well I’d rather fight the user fight than the investor fight. That was the case for us. It was so interesting. We never had a better strategy, we never had a better plan, we never had better execution than two moments. One, when we were running out of money and two, when we were hit break even. It was just wow. The clarity was amazing. The fear went away. It’s like, this isn’t going to die tomorrow.
Craig Cannon [54:49] – Well, you have confidence and then all of a sudden, when you look at the other side of the table, now you have a product that the investor wants.
Michael Seibel [54:56] – Isn’t that funny? When you don’t need investment, guess who comes around?
Craig Cannon [55:01] – Oh you need some more money, yeah no problem. I’ll cover you guys. Classic. Basically the way you wrap up is this is about leverage. What you say is, “Don’t limp into a Series A fundraiser. You need to be able to show that you have taken the early investment money and used it sensibly to create a product that people love. You need to have sustained growth to raise a Series A. Understand that and you’ll be better off than most start-ups.”
Michael Seibel [55:28] – It’s kind of simple, right? It’s interesting, we have a Series A program at YC and we help YC companies, tends to be like 12 to 24 months after YC, Prop four and raise Series A. At the kick-off meeting, it’s very similar to the kick-off meeting at YC. Everyone goes around in circles, that’s what they do. But the difference is everyone says their revenue too. It was so much fun to be at the kickoff meeting for the Series A because you’d hear some ideas like, “Oh I’m doing yada, yada, yada,” and like, Yeah at YC we hear ideas all the time. It’s like yeah okay send them an idea.
Craig Cannon [56:08] – Seems cool.
Michael Seibel [56:09] – Yeah, and then the founders like, yeah and we’re doing $5,000,000 revenue and you’re like oh, that’s different. And the next person’s like, yeah and we’re doing $3,500,000 revenue, and the next person’s like oh we’re doing $4,000,000, and it’s like, so funny how an idea sounds better when there’s a revenue number after it. What’s so funny is that, I think what YC’s good at is if you can create the business leverage. If you can make your business work, we can help you present that in the leverage maximizing way and we can help you do a process that maximizes the leverage that you’ve created. But you have to create the core leverage. You’ve got to do the work. You got to figure out the product that gets that usage and then we can help you package it and sell it most effectively. It’s fun to see when that works. You can’t predict it. If you ask me years before who was going to be in that Series A program, I would not have known. But man, those founders are far more formidable. When they have something, they’re killers. Quiet strength when you got shit —
Craig Cannon [57:20] – You don’t need to flaunt it.
Michael Seibel [57:21] – No! It’s funny ’cause sometimes I see people with snazzy pitches and snazzy, things that remind me of stuff you would hear on a TV infomercial and you don’t need that much work when you have a good company. Usually you can just show graphs and it’s like “Oh, wow that’s really working.”
Craig Cannon [57:47] – Awesome, alright thanks man.
Michael Seibel [57:48] – Thank you.
Craig Cannon [57:51] – Alright, thanks for listening. As always, you can find the transcript and video at blog.ycombinator.com and if you have a second, it would be awesome to give us a rating and review wherever you find your podcasts. See you next time.