- The speaker and his co-founder started their first company, Kiko Calendar, in 2005 but lacked the necessary skills and knowledge for success.
- They made repeated mistakes and failures with various startup ideas, including building a calendar similar to Google Calendar.
- They eventually sold Kiko Calendar on eBay and raised more money to launch a reality television show called Justin.tv.
- They faced challenges with producing reality television and decided to open up the Justin.tv platform for anyone to broadcast live video.
- Despite initial failures, they managed to keep going and eventually found some success and growth.
- They struggled with understanding what was driving their growth and were unable to sustain it.
- They focused on monetizing their site and cutting costs to stabilize the company.
- They then faced a decision on what to do next and explored opportunities in mobile video and gaming.
- They set goals and formed teams to work on these projects, eventually leading to the spin-off of Social Cam and the speaker's focus on gaming.

- The founders believed in the potential of gaming content and decided to launch Twitch.
- They had limited market research but felt confident in their ability to talk to users.
- They built products for users and gradually improved their understanding of their needs.
- They focused on building a platform for broadcasters and gamers.
- They went through the typical ups and downs of a startup, but eventually found success.
- They advise not to give up, take care of mental and physical health, and stick with it to succeed in startups.

Emmett Shear at Startup School Silicon Valley 2014

View the annotated transcript, and add annotations of your own, on Genius: http://tech.genius.com/Emmett-shear-emmett-shear-at-startup-school-sv-2014-annotated

We packed ourselves into my Civic and we drove up into Boston where we'd just been accepted, thankfully, into the Summer Founders Program.

Our backup plan was to go live with our parents, So that was great.

Both Justin and I were very relieved that we could get our own place, albeit a little tiny bedroom in Roxbury, which is all we could afford at the time.

I still to this day don't quite know what they saw in us to accept us, because looking back on it, we knew none of the things we needed in order to succeed in startups.

And we missed my first slide, which is a big slide that says, imagine in front of you, don't give up.

Because that's the only important thing that I really had going in when we started here was a determination not to give up.

Because if you look back on it, we kind of knew how to program, we weren't very good in retrospect, we didn't really know what we were doing.

I thought that you had to use a database from Java, I didn't understand that that was like a separate thing and that they had nothing to do with each other.

We really, really didn't know what we were doing.

We didn't know how to make product decisions.

We certainly didn't know how to manage people.

All we had was a desire to make it happen and an amazing support network of people who were helping us make it happen.

And so we started our company Kiko Calendar.

And Kiko Calendar was a story in just repeated mistake and failure.

We were building basically Google Calendar.

I wasn't called that at the time because Google Calendar hadn't been released, but we'd seen Gmail as had almost everyone else, and we and about 9 other companies all had the same brilliant idea at the same time, which is We should make a calendar that works like Gmail, which is actually a pretty good idea, it turns out, but everyone had that same idea, so it probably wasn't a great start at the time.

Anyway, so not only did we have a kind of not so great idea, but we couldn't stay focused on it for more than like 5 weeks in a row, so we had like this founder ADD, we'd go and we'd work on Kiko for about 5 weeks, and it wouldn't explode overnight, mysteriously for some reason, and we'd get demoralized and we'd give up and we'd go build and launch some other product, like a social network for families, or a search engine for MySpace.

MySpace was big at the time, not Facebook, so just context.

We built something called Sounds App, which was an imagined sound cloud, but built by people who didn't have any idea or vision for how to grow it from there.

We did get a really great education in prototyping products.

We built and launched no less than 6 startup ideas in a year and a half.

And we kept going back to Kiko when that inevitably also failed to explode.

We learned a lot about how ADD definitely prevents you from succeeding.

I was much better by the end of that than I was at the beginning, as was Justin.

We hired our first employees, screwed up hiring our first employees and managing them, but learning experience.

And then we sold it on eBay, because Google Calendar had been launched, and we had no idea how to compete with that.

So we knew after we did that we wanted to keep doing startups.

We sold it on eBay for about a quarter million dollars, which at the time was like an inconceivably large amount of money.

And we raised more money from YC with this basic concept that we were going to build a reality television show around Justin's life.

The Decision making by YC, looking back now, looks good, but at the time, I can't really understand what they were thinking.

So we knew at the time, we knew 2 things.

We knew we were missing 2 really crucial skills in order to make a startup happen.

We needed someone who could keep us from just meandering to build a third start-up.

So we brought on our friend from college, Michael Seibel, to be the CEO and be our parent, sort of.

Keep us product-focused guys from getting ADD and leaving the project and keep us on the straight and narrow.

We promised him we would move to New York really quick.

He should come to Silicon Valley for us to launch it.

We will raise money and move to New York.

I think we kind of intended to, but not really.

And so then we brought on Kyle, our fourth co-founder, because we needed someone who actually knew how to build hardware.

We thought we were smart, but we were like kind of junior-esque programmers at the time and definitely couldn't build what we thought we needed to build for Justin.tv to launch it, to launch a live 24-7 reality television show.

This was our awesome four-person founding team.

Four-person founding teams are generally a bad idea.

But something about the way the team dynamics gelled made it function and we launched the show.

And It was our first time launching a show where we got the attention of the whole country.

I mean, maybe not everybody, but a lot of people heard about us.

We were on Ann Curry on the morning show.

We built something that people outside of Silicon Valley cared about.

We had hit like 150,000 uniques in the first month.

And then it turns out we have no idea what we were doing in terms of producing reality television.

And so We went back to what we knew about, which was the technology and the platform, the thing we'd spent the previous year and a half working on, and we opened up the Justin.tv platform to everyone and made it into the website you may have actually visited at some point, which was a platform for anyone to broadcast live video.

So you can see Justin.tv being flat for a very long time as we had no idea what we were doing and bumbled around with reality television shows, you can see that bend in the curve.

That's us figuring out, oh yeah, we should probably just not produce the content and let other people do it.

It was really important that we didn't give up, right?

So the the key thing that happened there is we had a failed project and a failed startup.

But rather than give up on it, we said how do we repurpose this, how do we keep going, we managed to raise money despite the fact that it was not a good idea.

Before we opened the platform, people believed in us and gave us money.

There's a current theme here, we convinced people to believe in us and give us money before we actually had figured it out.

And that was really, really good for us because it let us keep going and not give up.

But we would have kept going anyways, I think.

Even if we hadn't been able to draw a salary, which we couldn't for months in the middle, we would have kept going.

We got product market fit, or we sort of thought we did.

We'd stumbled accidentally into something people really wanted.

And so then you get to this period in Justin.tv's history where you have some growth, and everyone's really excited.

I had no idea how to run a web service that got more than 3 users at a time.

And so we had a lot of downtime, a lot of downtime.

Kept growing anyways, and got a lot better at it learned how to scale a web service and then The Nick the sad thing happened, which is that we didn't understand what was Generating our growth And so we couldn't cause it to keep going.

We hit some point and stopped growing, and it was totally mysterious to us why this happened.

We didn't really, if you'd asked us, why were you growing 3 months ago and you're not growing now, we would not have been able to tell you, because we couldn't have told you why we started growing in the first place, other than we made some changes and it seemed like it worked.

So we kept going, we kept going for years, we realized, okay, well, we're not growing.

It's 2008, so if you cast your mind back, the market has totally collapsed, there's no way we're going to raise money.

And we were like, well, we better start making some money.

And so we worked really hard on monetizing our site, we cut costs, and we clawed our way, I wouldn't call it profitability, but at least we'd staunched the bleeding.

And Then we kind of found ourselves stuck in this place where, what do we do now?

Like, we've got this thing, we could stay here, we haven't been able to make it profitable, we could work on it for a while, you know, it's not really going anywhere.

And so we had this moment where we all met up and we're like, we need to do something with this company.

And there's sort of 2 schools of thought.

At the time, the iPhone was new, and we thought there's this real opportunity to build mobile video.

There was no good mobile video players in the space, and we thought that could be big.

And there was this sort of second idea around gaming, Which I was interested in primarily because it was the only content on Justin TV that I personally watched and We actually said okay this time rather than just launch random features and see But I don't know just sort of guess whether they worked or not.

We're gonna set real goals for 2 projects, 1 around mobile, 1 around gaming, and we're going to try to hit those goals.

And so we set up 2 teams internally, 1 led by Michael, our then CEO, which turned into Social Cam, wound up getting spun off.

And then a second 1 led by me in partnership with Kevin Lin, who was our COO.

And The 2 of us really believed in the gaming part mostly because we're big gamers.

And I'd done some hand-waving math about how IGN was big, so maybe we could be too.

And we'd looked into it, and said, okay, There's no really content issues with this.

It's not incredibly expensive the way sports would be.

And we realized that advertisers were OK with gaming.

That basically was, in fact, the extent of the market research.

And so that was this project that was codenamed Zarth.

Anyone who's ever worked with me knows that I love codenames.

I will codename like moving the kitchen, anything.

So our project codename for this was Zarth, which no 1 else in the company loved, but I really loved.

I wanted to launch Twitch as Zarth, but they wouldn't let me, so we had to come up with a new name.

The best thing about the work we did then was that that was when the light bulb went off.

That was when I finally sort of I guess made the final step, this is 5, 6 years in as an entrepreneur, and got the, had put together all the skills I needed to actually run a company.

I could build products, I understood the engineering side, I could manage people, I'd spent years painfully making mistakes, losing lots of employees.

Sorry Eric, sorry Tim, you were learning experiences.

And we, I'd figured out how to do, you know, hiring, we figured out how to scale products if we needed to.

We knew, I knew about how companies might make money, might lose money.

And What I finally could do when we were here is I finally got the extra bonus skill of figuring out how to talk to users.

So now you're all thinking I've heard about it, I have to talk to users, yeah, yeah, yeah, of course I talk to users.

And I thought I talked to users before we did Twitch.

I didn't actually understand how to talk to users and 80% of you when you go start your startup will make the exact same mistake I did which is you go talk to some users and you'll think you've done it right.

It's really hard It's something that we you know, We train product managers on when they come into the company.

It's something that I think is really 1 of the key skills for a startup founder.

It certainly was for me 1 of the missing skills.

And so we finally built things for our users.

It was like we'd been bumbling around in the dark trying to get across the room.

If you got across the room, that's product-market fit.

We'd be like, aha, we know, it's that way.

And then we'd go, oh, ah, stepped on a Lego, bumped my foot into a chair.

That's like, that was what building stuff was like.

And you can actually make progress that way.

Every time you hit something, you back up and you go around and you hit something else, you back up, you go around.

And eventually you get closer and closer to the direction you want to go in.

It was like I just turned the light switch on.

Oh, I see, there's some Legos there and the chair's there and I should just walk in this path and I'll get across the room.

And you put all of those things, accumulation of 6 years of experience together of basically more or less failing to do the right thing in a startup together, And you finally got Twitch TV, which we launched at E3 in 2011, about 9 months after the project started and about 4 months after I had this epiphany.

My designer is probably very, it's not bad.

My designer is very unhappy I put it up here because he's not happy with it because I made him do it in like 24 hours because I was like, oh right, we need a logo, we're launching in 24 hours, make it.

And so he did, and I actually think he did a pretty good job for a 24 hour logo.

And that was the whole thing of Twitch, was like we basically just threw these 2 or 3 things together.

We didn't do that much work compared to the amount of work we'd done for Justin.tv, but we applied it in the right places, in the right directions, because we finally knew what we were doing well enough to apply our effort efficiently.

And actually from that point on, the story gets way more boring.

Because we basically did the same thing over and over and over again.

We went and talked to the users, we'd ask them, like, what do you want?

And we'd get to know what their, not just what they wanted, but what their life goals were, and their experiences they have today, how they made money today, what their jobs were like, what they wanted to do.

And we identified, really importantly, which users were the most important to us, and that was the broadcasters.

We had to build a place that was amazing for people to broadcast gaming video, and we had to build a place that was amazing for gamers.

We would talk to them, we'd go build stuff, some of them would switch, we'd get enough leverage through promotion that we could bring more people onto the platform.

We started being able to do business development deals because we got bigger.

If you're starting a company, it's useless.

You see big companies do it or even medium-sized startups do it sometimes.

Every time we tried to do a business development deal during Justin.tv and we didn't have a product that was taking off, it was a gigantic distraction and waste of time.

It's only a bit beneficial once you get later on into the process.

At any rate, so you've probably seen this graph before.

I think this graph was actually inspired by Justin.TV, because this is what it feels like doing a startup, and this is what Justin.TV felt like the whole time, which is this rush of excitement in the beginning as you're first building something that you are really excited about, you have this idea, it's awesome, you're taking it to the world.

This long period of pain where it doesn't work and you don't know what you're doing and you are probably screwing it up.

If everyone was super skilled and knew everything they needed to know from day 1, they might be able to enter that growth curve immediately.

You have to have massive learning experience.

And learning experience is specific to every startup.

That's why I don't really think giving generic startup advice usually works.

Because even that business development advice I just gave, that's wrong.

Some number of people, that's actually completely the wrong advice.

You should actually go do biz dev deals before you even launch your company.

I don't know what sort of that would be for, but there's got to be 1 out there that fits that pattern.

Eventually if you stick with it, even if you have to change your idea, even if you have to pivot the whole company in a different direction, if you have to start over with a new product idea entirely, you get there.

Eventually you get something that is actually working.

And it's a reflection of you as much as it is of learning about the domain.

I mean, being a domain expert is valuable, but you also have to build the skills.

And I know when we started Justin.tv, and we started Kiko way back in 2005, we didn't have the skills, and it was only through sticking with it the entire time that we built them.

So don't give up, and if you just keep at it, you will get better.

And even if your first startup doesn't work, maybe your second 1 will.

I'm apparently a bit of a slow learner because the other people seem to have done it faster sometimes, but this was a great experience the whole time, and if you don't love it, you won't make it through the long period of pain that is inevitable.

So make sure that you take care of yourself during the process, make sure that you take care of your mental health, your physical health while you're doing it because it's a long road.

I think earlier they were saying it takes an average of 8 years, it took us 8 years, a gate and a half.

In fact, it could have easily taken longer.

The Amazon deal kind of came out of nowhere.

We weren't really looking to sell the company.

And so it could have been 9 years, 10 years before you saw any kind of exit or end of the road on that stuff.

And for me, it isn't even the end of the road.

I mean, I actually, after 8 years of this, I love it.

And for me, it's something I want to keep doing going forward indefinitely into the future.

And so that's really the most important thing is stick with it all the way and don't give up.

It's the only way to succeed in startups.


Speaker 0: Hello, everybody. Does it actually work? Yes. All right. This was us in 2005.

We were just out of college. We packed ourselves into my Civic and we drove up into Boston where we'd just been accepted, thankfully, into the Summer Founders Program. It's very first year. Our backup plan was to go live with our parents, So that was great. Both Justin and I were very relieved that we could get our own place, albeit a little tiny bedroom in Roxbury, which is all we could afford at the time.

And that summer was amazing. I still to this day don't quite know what they saw in us to accept us, because looking back on it, we knew none of the things we needed in order to succeed in startups. And we missed my first slide, which is a big slide that says, imagine in front of you, don't give up. Because that's the only important thing that I really had going in when we started here was a determination not to give up. Because if you look back on it, we kind of knew how to program, we weren't very good in retrospect, we didn't really know what we were doing.

I thought that you had to use a database from Java, I didn't understand that that was like a separate thing and that they had nothing to do with each other. We really, really didn't know what we were doing. We didn't know how to make product decisions. We didn't know how to raise money. We didn't know how to hire people.

We certainly didn't know how to manage people. And we had basically none of the skills. All we had was a desire to make it happen and an amazing support network of people who were helping us make it happen. And so we started our company Kiko Calendar. And Kiko Calendar was a story in just repeated mistake and failure.

We were building basically Google Calendar. I wasn't called that at the time because Google Calendar hadn't been released, but we'd seen Gmail as had almost everyone else, and we and about 9 other companies all had the same brilliant idea at the same time, which is We should make a calendar that works like Gmail, which is actually a pretty good idea, it turns out, but everyone had that same idea, so it probably wasn't a great start at the time. Anyway, so not only did we have a kind of not so great idea, but we couldn't stay focused on it for more than like 5 weeks in a row, so we had like this founder ADD, we'd go and we'd work on Kiko for about 5 weeks, and it wouldn't explode overnight, mysteriously for some reason, and we'd get demoralized and we'd give up and we'd go build and launch some other product, like a social network for families, or a search engine for MySpace. MySpace was big at the time, not Facebook, so just context. We built something called Sounds App, which was an imagined sound cloud, but built by people who didn't have any idea or vision for how to grow it from there.

We did get a really great education in prototyping products. We built and launched no less than 6 startup ideas in a year and a half. And we kept going back to Kiko when that inevitably also failed to explode. And so we learned a lot through that. We learned a lot about how ADD definitely prevents you from succeeding.

We learned about how to program. I was much better by the end of that than I was at the beginning, as was Justin. We hired our first employees, screwed up hiring our first employees and managing them, but learning experience. And then we sold it on eBay, because Google Calendar had been launched, and we had no idea how to compete with that. So that takes us to the next step.

So we knew after we did that we wanted to keep doing startups. We sold it on eBay for about a quarter million dollars, which at the time was like an inconceivably large amount of money. We didn't keep all of that. Our investors got most of it. But still, It seemed like sort of a win.

And we raised more money from YC with this basic concept that we were going to build a reality television show around Justin's life. Again, I'm not sure. The Decision making by YC, looking back now, looks good, but at the time, I can't really understand what they were thinking. So we knew at the time, we knew 2 things. We knew we were missing 2 really crucial skills in order to make a startup happen.

We needed someone who could keep us from just meandering to build a third start-up. So we brought on our friend from college, Michael Seibel, to be the CEO and be our parent, sort of. Keep us product-focused guys from getting ADD and leaving the project and keep us on the straight and narrow. We promised him we would move to New York really quick. It would be great.

He should come to Silicon Valley for us to launch it. We will raise money and move to New York. That was all lies. We never moved to New York. I think we kind of intended to, but not really.

And so then we brought on Kyle, our fourth co-founder, because we needed someone who actually knew how to build hardware. We thought we were smart, but we were like kind of junior-esque programmers at the time and definitely couldn't build what we thought we needed to build for Justin.tv to launch it, to launch a live 24-7 reality television show. She brought Kyle in for that. This was our awesome four-person founding team. Four-person founding teams are generally a bad idea.

Ours worked. I don't generally recommend it. But something about the way the team dynamics gelled made it function and we launched the show. And It was our first time launching a show where we got the attention of the whole country. I mean, maybe not everybody, but a lot of people heard about us.

We were on Ann Curry on the morning show. We built something that people outside of Silicon Valley cared about. And that was really, really cool. And we had this awesome spike of growth. We had hit like 150,000 uniques in the first month.

And then it turns out we have no idea what we were doing in terms of producing reality television. We did not understand that. And so We went back to what we knew about, which was the technology and the platform, the thing we'd spent the previous year and a half working on, and we opened up the Justin.tv platform to everyone and made it into the website you may have actually visited at some point, which was a platform for anyone to broadcast live video. And so the... Come on.

There we go. Okay. So you can see Justin.tv being flat for a very long time as we had no idea what we were doing and bumbled around with reality television shows, you can see that bend in the curve. That's us figuring out, oh yeah, we should probably just not produce the content and let other people do it. It was really important that we didn't give up, right?

So the the key thing that happened there is we had a failed project and a failed startup. Like it wasn't working. We were, it was clearly going nowhere. But rather than give up on it, we said how do we repurpose this, how do we keep going, we managed to raise money despite the fact that it was not a good idea. Before we opened the platform, people believed in us and gave us money.

There's a current theme here, we convinced people to believe in us and give us money before we actually had figured it out. And that was really, really good for us because it let us keep going and not give up. But we would have kept going anyways, I think. We were all really committed. Even if we hadn't been able to draw a salary, which we couldn't for months in the middle, we would have kept going.

So then it's great. We got product market fit, or we sort of thought we did. We didn't really. But we'd hit something. We'd stumbled accidentally into something people really wanted.

And the growth sort of speaks to that. And so then you get to this period in Justin.tv's history where you have some growth, and everyone's really excited. And it's really just heads down. I've never scaled anything before. So this is another skill we didn't have.

I had no idea how to run a web service that got more than 3 users at a time. And neither did Kyle. And so we had a lot of downtime, a lot of downtime. Kept growing anyways, and got a lot better at it learned how to scale a web service and then The Nick the sad thing happened, which is that we didn't understand what was Generating our growth And so we couldn't cause it to keep going. We hit some point and stopped growing, and it was totally mysterious to us why this happened.

We didn't really, if you'd asked us, why were you growing 3 months ago and you're not growing now, we would not have been able to tell you, because we couldn't have told you why we started growing in the first place, other than we made some changes and it seemed like it worked. So we kept going, we kept going for years, we realized, okay, well, we're not growing. It's 2008, so if you cast your mind back, the market has totally collapsed, there's no way we're going to raise money. And we were like, well, we better start making some money. And so we worked really hard on monetizing our site, we cut costs, and we clawed our way, I wouldn't call it profitability, but at least we'd staunched the bleeding.

And Then we kind of found ourselves stuck in this place where, what do we do now? Like, we've got this thing, we could stay here, we haven't been able to make it profitable, we could work on it for a while, you know, it's not really going anywhere. And so we had this moment where we all met up and we're like, we need to do something with this company. And there's sort of 2 schools of thought. 1 of them was mobile.

At the time, the iPhone was new, and we thought there's this real opportunity to build mobile video. There was no good mobile video players in the space, and we thought that could be big. And there was this sort of second idea around gaming, Which I was interested in primarily because it was the only content on Justin TV that I personally watched and We actually said okay this time rather than just launch random features and see But I don't know just sort of guess whether they worked or not. We're gonna set real goals for 2 projects, 1 around mobile, 1 around gaming, and we're going to try to hit those goals. And so we set up 2 teams internally, 1 led by Michael, our then CEO, which turned into Social Cam, wound up getting spun off.

And then a second 1 led by me in partnership with Kevin Lin, who was our COO. And The 2 of us really believed in the gaming part mostly because we're big gamers. And I'd done some hand-waving math about how IGN was big, so maybe we could be too. And we'd looked into it, and said, okay, There's no really content issues with this. We can go get the gaming content.

It's not incredibly expensive the way sports would be. And we realized that advertisers were OK with gaming. We were like, OK, great. All the market research is done. Let's go do it.

That basically was, in fact, the extent of the market research. And so that was this project that was codenamed Zarth. Anyone who's ever worked with me knows that I love codenames. I will codename anything. I'll codename reorgs.

I will codename like moving the kitchen, anything. I love codenames. So our project codename for this was Zarth, which no 1 else in the company loved, but I really loved. I wanted to launch Twitch as Zarth, but they wouldn't let me, so we had to come up with a new name. The best thing about the work we did then was that that was when the light bulb went off.

That was when I finally sort of I guess made the final step, this is 5, 6 years in as an entrepreneur, and got the, had put together all the skills I needed to actually run a company. I could build products, I understood the engineering side, I could manage people, I'd spent years painfully making mistakes, losing lots of employees. Sorry Eric, sorry Tim, you were learning experiences. And we, I'd figured out how to do, you know, hiring, we figured out how to scale products if we needed to. We knew, I knew about how companies might make money, might lose money.

And What I finally could do when we were here is I finally got the extra bonus skill of figuring out how to talk to users. So now you're all thinking I've heard about it, I have to talk to users, yeah, yeah, yeah, of course I talk to users. And I thought I talked to users before we did Twitch. I was wrong. I didn't actually understand how to talk to users and 80% of you when you go start your startup will make the exact same mistake I did which is you go talk to some users and you'll think you've done it right.

It's really hard It's something that we you know, We train product managers on when they come into the company. It's something that I think is really 1 of the key skills for a startup founder. It certainly was for me 1 of the missing skills. And so we finally built things for our users. It was like we'd been bumbling around in the dark trying to get across the room.

If you got across the room, that's product-market fit. We'd be like, aha, we know, it's that way. And then we'd go, oh, ah, stepped on a Lego, bumped my foot into a chair. That's like, that was what building stuff was like. And you can actually make progress that way.

Every time you hit something, you back up and you go around and you hit something else, you back up, you go around. And eventually you get closer and closer to the direction you want to go in. It was like I just turned the light switch on. Oh, I see, there's some Legos there and the chair's there and I should just walk in this path and I'll get across the room. And it was really a magical experience.

And you put all of those things, accumulation of 6 years of experience together of basically more or less failing to do the right thing in a startup together, And you finally got Twitch TV, which we launched at E3 in 2011, about 9 months after the project started and about 4 months after I had this epiphany. And this logo is not very good. My designer is probably very, it's not bad. My designer is very unhappy I put it up here because he's not happy with it because I made him do it in like 24 hours because I was like, oh right, we need a logo, we're launching in 24 hours, make it. And so he did, and I actually think he did a pretty good job for a 24 hour logo.

And that was the whole thing of Twitch, was like we basically just threw these 2 or 3 things together. We didn't do that much work compared to the amount of work we'd done for Justin.tv, but we applied it in the right places, in the right directions, because we finally knew what we were doing well enough to apply our effort efficiently. And that was Twitch. And actually from that point on, the story gets way more boring. Because we basically did the same thing over and over and over again.

We went and talked to the users, we'd ask them, like, what do you want? And we'd get to know what their, not just what they wanted, but what their life goals were, and their experiences they have today, how they made money today, what their jobs were like, what they wanted to do. And we identified, really importantly, which users were the most important to us, and that was the broadcasters. We had to build a place that was amazing for people to broadcast gaming video, and we had to build a place that was amazing for gamers. And so, we would just cycle that.

We would talk to them, we'd go build stuff, some of them would switch, we'd get enough leverage through promotion that we could bring more people onto the platform. We started being able to do business development deals because we got bigger. Don't do business development deals. If you're starting a company, it's useless. You see big companies do it or even medium-sized startups do it sometimes.

It can be quite effective later. Don't imitate them. It doesn't work early. Every time we tried to do a business development deal during Justin.tv and we didn't have a product that was taking off, it was a gigantic distraction and waste of time. It's only a bit beneficial once you get later on into the process.

At any rate, so you've probably seen this graph before. I think this graph was actually inspired by Justin.TV, because this is what it feels like doing a startup, and this is what Justin.TV felt like the whole time, which is this rush of excitement in the beginning as you're first building something that you are really excited about, you have this idea, it's awesome, you're taking it to the world. This long period of pain where it doesn't work and you don't know what you're doing and you are probably screwing it up. If everyone was super skilled and knew everything they needed to know from day 1, they might be able to enter that growth curve immediately. But you just can't.

You have to have massive learning experience. And learning experience is specific to every startup. That's why I don't really think giving generic startup advice usually works. Because even that business development advice I just gave, that's wrong. Some number of people, that's actually completely the wrong advice.

You should actually go do biz dev deals before you even launch your company. I don't know what sort of that would be for, but there's got to be 1 out there that fits that pattern. And so eventually it gets better. I think that's the real message. Eventually if you stick with it, even if you have to change your idea, even if you have to pivot the whole company in a different direction, if you have to start over with a new product idea entirely, you get there.

Eventually you get something that is actually working. And it's a reflection of you as much as it is of learning about the domain. I mean, being a domain expert is valuable, but you also have to build the skills. And I know when we started Justin.tv, and we started Kiko way back in 2005, we didn't have the skills, and it was only through sticking with it the entire time that we built them. So don't give up, and if you just keep at it, you will get better.

And even if your first startup doesn't work, maybe your second 1 will. In my case, it took 3 tries. I'm apparently a bit of a slow learner because the other people seem to have done it faster sometimes, but this was a great experience the whole time, and if you don't love it, you won't make it through the long period of pain that is inevitable. So make sure that you take care of yourself during the process, make sure that you take care of your mental health, your physical health while you're doing it because it's a long road. I think earlier they were saying it takes an average of 8 years, it took us 8 years, a gate and a half.

And that's normal. In fact, it could have easily taken longer. The Amazon deal kind of came out of nowhere. We weren't really looking to sell the company. We weren't expecting to.

And so it could have been 9 years, 10 years before you saw any kind of exit or end of the road on that stuff. And for me, it isn't even the end of the road. I mean, I actually, after 8 years of this, I love it. I like really like doing it. And for me, it's something I want to keep doing going forward indefinitely into the future.

I really enjoy it. And so that's really the most important thing is stick with it all the way and don't give up. And that's the most important thing. It's the only way to succeed in startups. Thank you.

Thank you.

See all Y Combinator transcripts on Youtube

Emmett Shear at Startup School SV 2014