Episode 16: Justin Kan of Twitch and Atrium: From Builder to Entrepreneur

GGV Capital’s Hans Tung and Zara Zhang interview serial entrepreneur Justin Kan. In 2007, he co-founded Justin.TV, a website that allowed anyone to broadcast video online. In 2011, Justin.tv spinned off its gaming division as Twitch, which went on to become the leading live streaming platform for video games in the US. Twitch was acquired by Amazon in 2014 for almost a billion dollars. More recently, Justin co-founded Atrium LTS—the “LTS” stands for “Legal Technology Services” – a startup that’s building technology to revolutionize the legal industry. GGV is an investor in Atrium.

Justin is a true startup veteran—in addition to starting multiple companies of his own, he has worked with hundreds of startups as a partner at Y Combinator, and has also personally angel invested in over 65 companies. Justin grew up in Seattle as a second-generation Chinese American, and graduated from Yale in 2005 with degrees in physics and philosophy.

In this episode, Justin recounted the story of how he started to live stream his life before streaming became cool, the pitfalls he has gone through during his startup journey, whether it was the right decision to sell Twitch to Amazon in 2014, and what gets him excited about his new venture Atrium.

Join our listeners’ community via WeChat/Slack at 996.ggvc.com/community. GGV Capital also produces a biweekly email newsletter in English, also called “996,” which has a roundup of the week’s most important happenings in tech in China. Subscribe at 996.ggvc.com.

The 996 Podcast is brought to you by GGV Capital, a multi-stage venture capital firm based in Silicon Valley, Shanghai, and Beijing. We have been partnering with leading technology entrepreneurs for the past 18 years from seed to pre-IPO. With $3.8 billion in capital under management across eight funds, GGV invests in globally minded entrepreneurs in consumer internet, e-commerce, frontier tech, and enterprise. GGV has invested in over 280 companies, with 30 companies valued at over $1 billion. Portfolio companies include Airbnb, Alibaba, Bytedance (Toutiao), Ctrip, Didi Chuxing, DOMO, Hashicorp, Hellobike, Houzz, Keep, Musical.ly, Slack, Square, Wish, Xiaohongshu, YY, and others. Find out more at ggvc.com.


ZARA ZHANG: Hey everyone. Before we begin I just wanted to remind you that we have created a WeChat group and a Slack channel for you, our listeners. This is a community of people brought together by their shared interest in tech in China. Through these groups our listeners have made friends, recruited team members, and found clients.

We also organize regular offline meet-ups for this community. We had a meet-up in Beijing last month and we’ll be having another one in San Francisco on Thursday, July 26th. Details for these meet ups will be announced in our groups. To join the groups, please go to 996.ggvc.com/community and follow the instructions there. Once again that’s 996.ggvc.com/community.

HANS TUNG: Hi there. Welcome to the 996 podcast brought to you by GGV Capital. On this show we interview movers and shakers of China’s tech industry as well as tech leaders who have a U.S.-China crossborder perspective. My name is Hans Tung. I’m a managing partner at GGV Capital and I’ve been working on startups and investing in them in both the U.S. and China for the past 20 years.

ZARA ZHANG: My name is Zara Zhang. I’m an investment analyst at GGV capital and a former journalist. Why is this show called 996? 996 is the work schedule that many Chinese founders have organically adopted. That is, 9 a.m. to 9 p.m., 6 days a week.

HANS TUNG: To us, 996 captures the intensity, drive and speed of Chinese internet companies, many of which are moving faster than even their American counterparts.

On the show today we have Justin Kan. In 2007, he cofounded Justin.tv, a website that allows anyone to broadcast video online. In 2011, Justin.tv spun off its gaming division as Twitch, which as many people know, went on to become the leading live streaming platform for video games in the U.S. and elsewhere, which was acquired by Amazon in 2014 for about $1 billion. More recently, Justin cofounded Atrium LTS. LTS stands for Legal Technology Services. It’s a startup that’s building technology to revolutionize the legal industry and GGV fortunately is an investor in Atrium as well.

ZARA ZHANG: Justin is as a true startup veteran. In addition to starting multiple companies of his own, he’s also worked with hundreds of them as a partner at Y Combinator and has also personally angel invested in many of them. Justin grew up in Seattle as a Chinese American and graduated from Yale with degrees in physics and philosophy. Welcome to the show, Justin.

JUSTIN KAN: Thanks for having me, guys. I am excited.

ZARA ZHANG: I’ll start with live streaming. I think you’re sort of the original live streamer. In 2007 you and your partners started Justin.tv, a 24/7 live video feed of your life, broadcast via webcam attached to your head. Your live casting lasted about 8 months and attracted a lot of attention. What inspired you to broadcast your life at that time when this concept didn’t even exist?

JUSTIN KAN: It was just a crazy idea. A random, pretty bad idea actually at the time. Although I think we were onto a nugget of something popular because you know Justin.tv was before Twitter, Instagram or Facebook being popular and it was before the Kardashians, before Instagram celebrities.

HANS TUNG: There were The Bachelor and that kind of show in the offline world.

JUSTIN KAN: Yeah. There’s TV reality but there was no real reality, like I’m taking you along on my life and we thought that would be interesting. Taking someone along on this journey of your life could be interesting and then the attention could make somebody a celebrity. Unfortunately, I am not very entertaining.

And so I was a very poor test case. We should have gotten someone who was more of an actor, athlete, someone more attractive than me, maybe someone who was a musician – someone with some talents. But instead when people came to Justin.tv, which was a 24/7 live show about myself and my cofounders, they saw 4 guys sitting in an apartment programming.

ZARA ZHANG: Did you just go about your lives as usual? Did you change anything?

JUSTIN KAN: At first we tried to do more entertainment stuff, but it really fell back into mostly trying to keep our website up. And programming is not interesting. It’s interesting to do it, but it’s not really interesting to watch it because watching someone program is a lot like watching someone write an essay or watching someone watch a YouTube video or play a video game like Starcraft. The video of the person itself was not that entertaining. We tried to run around the city. Unfortunately there’s a lot of downtime.

Technology wasn’t quite there yet. It was right on the border. There was no iPhone when we tried to launch it and there was no fast mobile internet connection. We ended up stringing together multiple modems from different providers onto one computer and then send the video by multiplexing it over multiple cell phone connections. That’s how we were able to create this livestream back in 2007.

HANS TUNG: What was the first thing that hit? What was the first product-market fit that you just found with Justin.tv?

JUSTIN KAN: What happened was, we launched this show and we thought we were so cool. A bunch of people came because they thought, ‘This is a real-life Truman Show or Ed TV.’ Now, of course, it wouldn’t be so weird. But back then, people were like, ‘Why do you want to put a video of your life on the internet? That is super weird.’ So it was this big media phenomenon and then we got interviewed by The Today Show and MTV and all this stuff.

People would come and they’d show up and see us on our computers and to be like, ‘This is boring. What are you doing?’ Luckily, they would come and ask us, ‘How are you making a livestream on the Internet?’ because that was not technology that existed. So we built a platform like YouTube for live video. We launched that about 6 months later after we had launched a show and then that’s when it really took off.

HANS TUNG: That’s the product I used. I didn’t understand why it was called Justin.tv back then.

JUSTIN KAN: Right. A lot of people thought it was ‘Just In TV’, like live TV, but no, it was named after a guy named Justin. I’m a very modest guy. I’m a humble guy.

ZARA ZHANG: Did you have any idea that it was going to go on and become so big and launch a whole category or did you do it just for fun?

JUSTIN KAN: For the show itself, we thought it could be a crazy idea. Maybe it would be big, who knows. We had no idea how big it would be. But for the site – YouTube for live video – we thought that could be big. YouTube had already sold for $1.6-1.7 billion by that time. So there was an idea that video online could be valuable. So it wasn’t that much of a stretch once we pivoted to be a platform. Within a couple of years, we were at 20 million MAU on that platform.

HANS TUNG: That’s a great number – that was huge.

ZARA ZHANG: Can you tell us the story of how Twitch became a spinoff of Justin.tv?

JUSTIN KAN: We launched this site in October 2007. Technically, it was a pretty big achievement for us because nothing worked back then: the flash media server didn’t work, streaming was much more difficult, people had to install plug-ins and stuff like that. So we launched the site and then it kind of took off for the next couple of years between the end of 2007 to early 2008 all the way through around 2010.

It had done really well and celebrities started live streaming. There was other all sorts of different content: sports, entertainment, talk shows. Then it just stopped growing because it turns out that actually, not all content is good live. Only very few kinds of content are good live and most content you want is VOD – a produced video.

So we needed to figure something out because anything that’s not growing on the internet is about to fall off a cliff. Because the internet is growing. So we decided to go back to the drawing board and figure out something that we think we can invest in that could potentially be bigger than what we have right now.

We looked at what we were doing and there were a couple ideas we had. One was around mobile video – kind of like Periscope actually – but 4 years earlier than Periscope. Then the other was the gaming channel. There was a small gaming section on our site. 3% of the site was people watching other people play video games. We were like, ‘Why are people doing this?’.

The interesting thing is that back when we had our own show 3.5 years earlier, my friend Steve Huffman – who’s a co-founder of Reddit – and I had been hanging out in the early days of the show and we plugged in the camera – an analog camera – into the Xbox and we were broadcasting ourselves playing Xbox. People liked that. Unfortunately we didn’t figure it out until way later.

In 2010, people started streaming gaming content and we decided to go with it and try to make this a thing. We think this can be bigger than the existing site because it seemed like engagement was high. One of my cofounders, Emmett, who’s the CEO of Twitch today, really thought this could be something. It was the type of content that he liked.

We decided to build a small team around focusing on the product here and investing in that. Long story short, after about 6 months of working on it, we said, ‘Hey this is really working.’ It was called The Justin.tv Gaming Channel at that point but then we changed the name to Twitch and launched it as its own site and that just started taking off.

HANS TUNG: How did you decide on Twitch as the name?

JUSTIN KAN: We were debating on the name. I remember I really liked some 4-letter acronym that was kind of like ESPN. It would have had X or G or TV or something like that. I don’t even remember what it was. There was absolutely no way we were going to do that. And then he had this idea of calling it Xarth. The domain was purchasable. I think we bought it for a couple thousand dollars. That was the placeholder name, the secret project name. So he was going to call it Xarth and then a few days before launch he found the site Twitch.TV. Twitch, like fast twitch muscles. It sounded like gaming. At that point, everyone had argued so much I think he was just like, ‘OK, whatever you want.’

ZARA ZHANG: How were you making money for your original live streaming product?

JUSTIN KAN: The original Justin.tv was very advertising-supported and it was all pre-roles but we also had this cool thing that were mid-roles that the broadcaster could trigger. So if you were doing a live podcast or something and you wanted to take a bathroom break, you could trigger advertisements and then go to the bathroom. We thought that was a huge innovation – which it was at the time. It’s still a thing today at Twitch, but now Twitch is more and more like the Chinese sites: it’s more direct user donations, virtual goods.

HANS TUNG: You guys have 4 cofounders. First of all, why so many, and two, how do you figure out how to make decisions?

JUSTIN KAN: That’s a great question. We made decisions incredibly poorly. Now that sounds like a joke but it’s not a joke. There’s probably a lot to talk about, in how that’s affected my subsequent companies or Atrium today, which had a lot of cofounders. Back then, we didn’t know anything. These are 4 22 year old guys who started the company as friends. We didn’t know anything about management. Nobody had ever worked at a real job before. Michael, who was the first CEO, had actually worked at a job but it was a political campaign and you can imagine that they don’t have the best management structure or training. So we have 4 guys and we got together and we made all of the important decisions by arguing among 4 people.

Emmett was the CTO. Kyle was the V.P. Engineering. Michael was the CEO. I was Chief Product Officer, but we made every important decision together. That is not a good way to do things. You need to trust. You need to delegate. You need to have division of labor. But we didn’t do that because we didn’t figure out how to do that until we pivoted to Twitch. We also incubated this other project called Social Cam. We couldn’t even decide what to pivot on. We pivoted to 2 things: Social Cam and Twitch. Social Cam eventually spun out. We sold it to Autodesk for $60 million. But that is ‘defies all odds’ type of thing.

HANS TUNG: Despite all the mistakes you made, you actually had a pretty good outcome. Two of them.

JUSTIN KAN: But this is not a good way to run things. So it took everybody going off on their own in order to have their own domain. Which is sad because we were in a lot of ways, a great team, but it’s not sad because everybody has done super well. My co-founder Kyle ended up starting a company with my brother, called Cruise Automation that is building self-driving cars.

HANS TUNG: I met him before he sold it to GM.

JUSTIN KAN: You should have invested!

HANS TUNG: Everyone thought it was a bit too early but Kyle is interesting.

JUSTIN KAN: They’ve done incredibly. They sold to GM and just raised $2 billion from Softbank. Incredible story. So truly, there was a lot of talent in this one organization but it took a little time to mature into like leadership.

ZARA ZHANG: Why did you still have so many cofounders for your current company if that didn’t work out before?

JUSTIN KAN: It did work out but now there’s a lot more delegation and ownership of specific roles. Our cofounders who’s the CTO is really in charge of the tech. I’m not micromanaging and saying ‘Let’s make this technical decision or this product decision.’ We have a co-founder who’s really in charge of the legal side. I don’t actually even know certain things about our legal, like who our clients are. I’m not managing at that level.

It’s much more about creating actual metrics of accountability and then saying ‘Go hit these metrics. I’m going to trust you. I’m going to give feedback, I’m going to be a resource to you if you need feedback on your plan. But I’m going to trust that you are handling this domain.’ We have a co-founder on the operations side and then one also on the marketing side.

HANS TUNG: We’ll talk in more detail about Atrium. Still on Twitch, Emmett is the CEO. When you guys decided to sell the company, $1 billion is an amazing outcome obviously. Did you ever look at how this sector is evolving in China and how virtual items have become such a great way to monetize. Did you ever think about how the site could have been more than what it is today?

JUSTIN KAN: I’ll tell you two things. The site obviously could be more and has become more. When we sold it, we were 55 million MAU and that’s a lot. But at the same time, at that level, it was very hard for us to raise capital because most investors at the time did not understand it. If we had talked to more probably Chinese investors, actually they would have understood it and in fact we had money from Chinese investors. But Sand Hill investors did not understand this industry. Of course one year later they all were investing in eSports or whatever.

But the point is nobody understood it so we thought $970 million was a tremendous amount for it. About 8 months ago, I tried to sell some secondary for $200 million and got a lot of declines. So we thought it was really good. Afterwards, obviously it’s become a much greater success since that.

In the year following the acquisition, all the metrics doubled. The company hasn’t released much recently in terms of metrics, but the company’s grown to almost 1,500 people. Now it’s based in San Francisco. They’re opening up their own 9-story building. It’s a tremendous success. I was just looking at Alexa for the first time in years and they’re the #12 website in America now. So obviously there was a lot of headroom but unfortunately, we weren’t able to really convince capital sources to fund it as an independent company. But you can’t be too sour about that.

One interesting conversation we were having during the process of selling in 2014 over the summer, was on a conference call among all the board members. Everybody said we should sell, this is a great deal. One of our board members, Chris Pike, formerly of Thrive and now doing his own rumored thing, was like ‘No, don’t sell.’ We were all silent on the call, like ‘Why? What do you think it could be worth?’ Because you think of this market as not that big of a market. We don’t know, but how big could it be? And he was like, ‘I don’t know. $3-4 billion? More?’ Everybody was just silent. Nobody wanted to be like, ‘You’re crazy.’ so everyone was just silent on the call. And then we just started talking about something else.

Chris was the only one who was right. It was fun though. Such an amazing adventure.

HANS TUNG: We’re an investor and my colleague Jenny Lee is on the board of YY which started as a game commenting site back in 2006. Today it is a $10 billion mobile video site for gamers and it’s just spun off something that’s live streaming for gaming. Huya, on its own, is also a $2 billion market cap company, both listed on Nasdaq. It’s hard to imagine in 2006 that as a game commenting site that’s $30 million and it’s just so hard to believe back then that it could be a mainstream thing. You can see today that you could be bigger than most of the sports activities.

JUSTIN KAN: Absolutely because everyone’s playing videogames out there.

HANS TUNG: My 10 year old daughter’s on it all the time. She spends a lot more time on gaming than I ever did on sports when I was her age. You see the time spent – it’s almost the #1 activity for kids now.

JUSTIN KAN: That’s probably not good, but…

HANS TUNG: I tell myself it’s good for her to be a good programmer or designer later.

JUSTIN KAN: I think the gaming helps with their analytical reasoning and stuff like that. I definitely think it did for me when I was a kid.

HANS TUNG: And teamwork.

ZARA ZHANG: Were you a hardcore gamer?

JUSTIN KAN: I was probably. I was playing Counterstrike and World of Warcraft and Warcraft 3 when I was in college. Probably spent a little bit too much time playing. I actually did lose a lot of time to playing games. But then once we had started Justin.tv, it actually went down a lot because business and start ups are kind of like its own game. A very stressful and horrible game to play sometimes.

ZARA ZHANG: What was it like to sell a company to Amazon? What was that process like for you?

JUSTIN KAN: I didn’t really do most of the work because by the time, Emmett’s the CEO, I’m just on the board. I think for them it was very stressful. For any acquisition process, it’s very stressful. I think Amazon Corp Dev is one of the most disciplined in Silicon Valley. I think a lot of other companies get caught up in hype, in ‘We need this for strategic reasons.’ Amazon, out of all the companies I’ve seen or through investments we’ve sold is the most disciplined.

HANS TUNG: Meaning that they won’g overpay. We have sold companies to Amazon before. All the things we learned from how Facebook acquired Instagram and WhatsApp because if it works out it’s a lot bigger than they can imagine. If it doesn’t work out, it doesn’t matter how much you pay.

JUSTIN KAN: There’s two types of companies that are good at acquisition: the ones that are really truly strategic and visionary, who are making visionary bets, and then there’s the type that are more really disciplined. That works too if you’re truly truly disciplined. I think the people who really struggle are the ones in the middle who are kind of half and half and then they lose important deals because they were too tight or they overpaid for things that don’t matter.

HANS TUNG: People forget that we’re about 3 billion internet users worldwide now. If something does work, it will be massive. If it doesn’t, it doesn’t really matter how much you pay because you wasted it already. And if it does work, you’ll make up for most of the losses.

JUSTIN KAN: Absolutely. You’re riding the world’s biggest wave right now.

ZARA ZHANG: A report last year found that there is now a bigger audience for game video than audiences of HBO, Netflix, ESPN, Hulu combined which is kind of crazy to think about. Did you imagine this would happen or were you surprised when game streaming took off?

JUSTIN KAN: We were surprised how fast it was growing, the game streaming, in the early years. It doesn’t surprise me though. Back in 2007, the gaming industry passed the movie industry. That was a big thing. All these entertainment people were like ‘What the fuck’ nobody saw that coming. OK, well when you do MTV Cribs, you see that every athlete has an Xbox. That makes sense to me.

I think it was pretty obvious that gaming video was going to be a huge thing. It was already the #2 category on YouTube by the time we started Twitch and then 1-2 years in, it was the #1 category on YouTube. So it was not a big surprise. It’s actually a surprise that it didn’t happen earlier.

HANS TUNG: In China at the same time – 2006, 2007 – gaming was already the #1 monetization method for internet. It was bigger than e-commerce, it was bigger than advertising. Shengda and NetEase were both churning out games and then Tencent got into it. Now, 10 years later you see Tencent everywhere from Riot Games to Fortnite to Supercell. It’s amazing to see how big the industry has become.

ZARA ZHANG: What was it like to have this whole startup journey as an Asian American? How aware were you of your identity and did that change things for you in any way?

JUSTIN KAN: That’s a good question. I grew up in Seattle and I didn’t speak Chinese at home. I went to a small private high school, University Preparatory Academy, with mostly white kids. I never really thought that much about identity and my ethnic identity. At Yale, I had joined the Chinese American student association – the club – and that was cool. But I think it is a little bit different from a lot of people who grew up maybe in California or the Bay Area or something where there’s a lot. In San Francisco, it’s like 33% Asian American or Chinese.

In the industry, it didn’t feel like – in my YC network or whatever – that there were not that many Asian American founders. Although I think actually if you look at it statistically, there weren’t that many – it was about 10% of YC, but it felt like it was 30% of YC or something like that. I didn’t really feel like it changed things very much.

HANS TUNG: What was your first reaction when you went to China?

JUSTIN KAN: Before the last time I went to China, it was 2002. So when I went to China in 2002, they were about to open Three Gorges Dam. I did the cruise down the thing and it was rural. I thought China was a developing world. I went back last year in December 2016 and I was like ‘Oh my god. This is crazy. This is different world altogether.’.

I feel that way a little bit because I’m not taking that many frequent trips to Asia. I’m probably going once a year but it’s a different place. I’ll go to Jakarta or Singapore or Seoul and it’s amazing how all of these places have better infrastructure than America, better taller buildings, more buildings, worse traffic unfortunately. The world has changed so much. I think it’s super interesting. I wish the climate in Asia were better – I’d consider moving.

HANS TUNG: We’ll get through it. London went through it in the 1950s. Tokyo went through it in the 1960s. It’ll get better but it takes time.

JUSTIN KAN: I was blown away. Also by the work ethic and innovation that’s happening. It’s pretty obvious now. If you came to America in 2014 – that’s when a lot of Chinese investors started investing in Silicon Valley – and people were wondering where this money is coming from, what’s going on.

And if you had told U.S. investors or founders that the innovation is actually coming in from China, they’d say ‘You’ve got to be kidding me.’ Now you see obviously the dockless transportation models coming here. A lot of AI innovation is happening in China. The balance of innovation is shifting pretty rapidly. You should be pretty scared if you are an American patriot. At the same time, we’ve taken a lot of things for granted. We take immigration of skilled labor for granted. We should be trying to get every single engineer from around the world to live here and to stay here. We should be giving incentives. Because just look at the numbers: the bell curves of the populations might look the same, but China has 1.3 billion people, American has 330 million people. It’s just not competitive.

HANS TUNG: The number of STEM grafts this year is 4.7 million for China, just under 600,000 for the U.S. Obviously quality matters, but over there, people are just as smart, work twice as hard and there are a lot more of them – you do the math.

JUSTIN KAN: You don’t have to be a math major to draw the rest of the curves and where they intersect. I just feel like people are going to be very surprised at how fast change happens, especially in the global balance of power.

HANS TUNG: On that note, a couple of things. One, we all say competition is good and when there’s real competition that’s when true colors come out. Do you really want to learn from competitors and get better? Or do you want to sit around and bitch and try to keep them down. That’s a decision that the U.S. has to make.

Secondly, if you look at the way that Chinese startups started, they try to turn everything into a math or engineering problem. What is the end goal? Where are we at? How do we go from A to B? Just work backwards: what has to happen for this to happen? So the time it takes to scale is actually faster in China because the end problem is clearly defined. How you guys started with Justin.tv and Twitch – you tried to innovate and try to get to something – is a lot more creative and anything could happen, but it takes more time.

JUSTIN KAN: Absolutely. I think there are merits of both sides and you probably want to figure out ways that you can combine both.

HANS TUNG: That’s part of the reason why I wanted to do this podcast and interview you as the first Chinese American on the show. To share more of different ways to scale, to come up with a problem to solve, how to get the right product-market-fit so that both sides can learn from each other. It’s an open platform: whoever learns faster from both sides will win. We hope there will be more of an American audience listening to this as well.

JUSTIN KAN: Absolutely. A lot of the criticism from when I started in Silicon Valley back in 2006 through 2010 or 2012, was that the Chinese were just cloning everything. Now I think people are clearly like ‘Oh that’s not the only thing.’ But even if you think that Chinese people are culturally less risk-taking, less innovative, there’s still 4x as many people. So you could be 1/4 as innovative. Only 1 in 4 people could be as innovative as an American person and you’d still be competitive. People should think about that.

ZARA ZHANG: How do you grow from someone who is very creative and cool and from a hacking culture, just trying to build things into a manager or someone who can run a team and run a company?

JUSTIN KAN: That’s a really good question. When I was young, I felt like experience was worthless. Because Silicon Valley has this myth, this legend, of: we just get a bunch of smart young people who are sleeping in a tent in their living room and they just make something amazing and then BAM – Snapchat, Twitter.

But the reality is, Twitter is a great example. Look at the scaling of Twitter. Its growth was massively stunted by the downtime for years and years. And what I realized over time, experience can be really valuable. Especially around management stuff. The thing that I didn’t realize when I was younger was in Silicon Valley, I thought you had to rethink everything or rebuild everything from first principles. But not everyone’s Elon Musk.

Mostly, you just needed to innovate on 1-2 things really well and have a 10x on 1-2 axes and then the rest of stuff, you should just do the best practice – whatever they did at Google or Facebook. You don’t have to innovate on your HR or recruiting, you just need to do the top level. That’s something that I didn’t really realize.

You can hire people who have experience and bring them in to do excellent work. We did that at Twitch actually, years later. 6 years later after we started, we brought in Jonathan Simpson-Bint as our Chief Revenue Officer. He had built gaming ad sales teams before for IGN – he was a co-founder for IGN. He knew what to do and just plug and play and boom, this guy already hires his sales team and then start ramping revenue.

So we started learning that you can hire these leaders who have experience who can execute. That’s a big thing to realize. If you’re young, you might have really fresh, innovative ideas that change the culture, but you can have other people with more experience help you execute.

HANS TUNG: How do you figure that out early?

JUSTIN KAN: I don’t know. If you figure it out early, that’s how you become super powerful. That’s how you become Evan from Snapchat. Or Zuck. Most people struggle with it and get a middle outcome.

ZARA ZHANG: You’ve seen and advised hundreds of startups. What’s the most common mistake that you see in founders and what do you tell them?

JUSTIN KAN: The most common mistake I see in founders is that they’re not focused enough. Companies always need to focus as much as possible. That’s super hard to do. I understand the temptation. I’m always tempted to go as broad as possible. I think focus is really important.

Second thing is that people give up way too easily. Now that startups are in fashion, it’s a job, like a kind of career. Lots of people try it and then they just give up after a year or whatever and go do something else. I don’t think that great companies are made that way.

HANS TUNG: They value their opportunity cost as too high. And therefore they think it’s not working out, they should do something else.

JUSTIN KAN: It kind of makes sense because in technology now, the salaries especially in the Bay Area are insane. So your opportunity cost may actually be high.

Other mistakes that people make? It depends. It depends on what kind of company you want and how much risk you’re willing to take. But a lot of people – once something is working, they go too slow. They don’t hire the best execs they can to help them scale. They’re scared to let go of things. They’re scared to overspend, which is reasonable in some ways, but it also can limit your growth.

HANS TUNG: I notice a lot of founders underestimate how important is to have their drive, their passion to be the guiding force of culture in the firm. They always worry if they hire somebody too experienced, their VCs will replace them with someone who’s more experienced. What a lot of founders don’t knows is that you can hire experienced hires. But they will never have the vision and the passion you do to continue to expand the business and pushing the envelope. So if you just focus on doing that well and pushing the executives you hire, you can build a much bigger company.

JUSTIN KAN: Absolutely. You must learn to manage and you must learn to scale. Very quickly, after your team is probably more than 5 people, the marginal effort of your work – your direct individual contributor work – like you programming, you doing sales, whatever it is is – is very low. Now it’s 20% of the company. If you have a team of 25 people, it’s 4% of the company.

If all of your effort in the entire year is 4% of the company, if you just spend your time affecting people’s work product, you only have to make people more than 4% more productive through your management in order to equal know how much individual contributor work you’d be doing. So your ability to manage people is your leverage. And I think people go into that mode too slowly. Once they have product-market-fit when something works.

ZARA ZHANG: Let’s move on to talk about Atrium, your current venture. Can you tell our audience more about what the company does?

JUSTIN KAN: Sure. Atrium came from my problems, my experiences as a founder with legal. I was an involuntary power user of corporate legal services. Every time you want to do any major business transaction in America, you pay lawyers: when you’re raising funds, if you’re doing M&A (we did big M&A obviously), litigation. For all of this, you need to hire lawyers. I loved the lawyers that I had who basically gave me expert advice based on their expert knowledge.

But there were a lot of experiences where I felt like it’s cookie cutter stuff, crank-turning work, and I wondered why they aren’t using more software to deliver services, to automate things, to streamline the flow of information. Lead law firms out there in America don’t even use project management software. They don’t use Trello or Asana. They don’t use CRM software for their clients.

I sat down with an attorney and I had been a client of their firm for 6 years with Twitch and she didn’t even know who I was. That’s weird because this is a $700 million a year business and you didn’t even know that I was a client. That’s weird because the level of professionalism compared to a normal enterprise sales organization is just not there. There’s a bunch of reasons the industry is like that.

Our goal with Atrium was to change that industry. We’re making it better for founders and we’re making it better for attorneys and paralegals as well. We want to make legal streamlined. Atrium is a top tier technology-enabled law firm that delivers legal services to founders. We make it streamlined, faster, more transparent and we make pricing much more upfront so you can know what you’re going to pay.

For our attorneys and paralegals who come from top firms, our goal is to make things easier for them and to automate all the low-level work so that they can focus on the high-level advising stuff, which is what people want to do. You didn’t go to law school to crank out signature matrixes or to put together the first version of a boilerplate doc.

HANS TUNG: And on top of that, lawyers know the cap table. Every time we do a fund raise, you can wait for shares to be issued and stuff, but the lawyers know exactly what’s going on the cap table and that’s super powerful. You can imagine other services can be rendered by having that information.

JUSTIN KAN: Absolutely. So what we’re doing is we really think about that we’re building as a platform where we take legal documents, we turn them into structured data so we can, for example, read all your financing docs, understand who are the owners of your company. And then from there, we can do things like we can produce and render a cap table. Or we can make that data accessible in other ways back to the business owner. And so I think that’s the super valuable thing that we’re doing and I’m pretty excited about it. It’s only been a year. We launched in June 2017 so it’s been exactly a year and things are going pretty good so far.

ZARA ZHANG: Last year you guys raised $2 million in your Series A from over 90 investors, GGV being one of them. Why did you decide to do such a party round with so many investors?

JUSTIN KAN: I just wanted to throw the biggest party biggest party ever. So a party round in Second Valley is kind of what happens now. I think the same is true in China. Someone will go and they’ll raise money from the dozens of investors, usually angels and maybe funds for their seed round.

We chose to do that because we wanted our investors to funnel us portfolio companies, business. It’s worked pretty well so far, so we’re pretty happy with the results. Another thing is legal is something that people don’t necessarily want to take a risk on. We didn’t want investors to say ‘Hey, don’t try this service.’ Even if they just had been familiarized with us, it helps.

ZARA ZHANG: What has been the progress so far in terms of providing the actual services?

JUSTIN KAN: We’ve done work for over 240 clients so far. We’ve helped those clients raise over $400 million in the past year.

HANS TUNG: How do you do that?

JUSTIN KAN: We do a couple of things. We obviously do the legal around their financing. We also have this bootcamp called Atrium Scale where we teach people how to fundraise. One of the things we realized through the course of operating Atrium is that a lot of founders come in and they want help with the process. You and I are very experienced because we’ve done it so many times. A lot of guys are coming in and they just don’t know what to do. You see more and more technology companies coming from outside the Bay Area now because of the cost of living is so high here, but they don’t necessary have the connections, they don’t know what to do. So we put together a program that’s a 2-day long bootcamp for free for founders to teach them how to raise money. We help them work on their deck and their narrative.

ZARA ZHANG: Are most of your clients first time founders?

JUSTIN KAN: No, many of them are multiple time founders or we have clients that are later stage companies, like 200 person companies and stuff like that. But we do have a lot of first time founder clients as well.

ZARA ZHANG: So there’s an education component too.

JUSTIN KAN: Yeah absolutely. One thing I learned from YC is, how did YC scale? They built a lot around education, around free content out there in the market, events and community. I think that one of the things that we can do here for Atrium being a platform for companies, if I think about Atrium’s mission, it’s really to help fast growing companies and provide them with services and applications that help them grow even faster. So how do we do that? Well I think education and community and teaching people how to run companies and information that they need is a way to do that.

HANS TUNG: So founders can go to YC first, once they get funded, they can come to Atrium for the next phase in their growth stage.

JUSTIN KAN: Absolutely. I would love to be able to help founders in that way.

ZARA ZHANG: It’s been 12 years since you started Justin.tv. What’s one thing that you wish you knew when you were just starting out?

HANS TUNG: Knowing what you know today.

JUSTIN KAN: Knowing what I know today, I would have just sold it to Facebook in 2006. Besides that, I think that when you have something that is a market category leader in a growing category, I think Ben Horowitz says it, don’t sell it.

HANS TUNG: Besides that.

JUSTIN KAN: Okay. Besides that, I think we could have been much more effective if we had implemented actual management in the early days of Justin.tv and going through Twitch and done a couple of basic things – maybe hired senior leaders. I know it sounds so cliche to be like, ‘Oh just hire the gray hairs.’ But having some more senior leadership, structuring the company where there’s more division of labor among the cofounders, where we were metrics driven, goals driven. We never were like on OKRs or anything like that. I think putting in some of these management tools.

HANS TUNG: So having workshops around that will be helpful for your founders then.

JUSTIN KAN: I should do that. For Atrium, we should have an OKRs workshop because I think it’s super important. Understanding that management isn’t a job right. It’s not something that’s supposed to make your life not fun as a startup. It is a tool for you to be more successful. It’s a collection of tools for you to be more successful. I always thought that management was a dirty word because we were supposed to just hack on something until it was like Facebook or whatever. We didn’t really realize how like the working world works. And I think that’s the benefit of actually having a job before you start a company, which we never had.

ZARA ZHANG: Hans, I’m curious on your perspective on what Justin said about if are a category leader you should hold onto it. Do you think that applies in China too? Or is that a different story.

HANS TUNG: I think in China, most people know that from early on, just so many sectors got disrupted so that if you have a category leader, they never, never, never sell. What has changed in China is now you see industry consolidation. So #1, #2 player fight for 3 years as hard as they can, raise as much money as they can, to figure out who’s better than figure out a way to merge so they can own more of the world. And the one who sold out ends up becoming a great angel investor. So now you have best of both worlds: you have more angel investors that are more professional, have experience going through that scaling experience and you also have a much bigger outcome in the original category. So it’s win-win.

ZARA ZHANG: We recently saw Mobike being acquired by Meituan and sometimes I guess the business is just hard to be a stand alone business depending on what you’re doing, it can be hard to be a stand alone business.

HANS TUNG: You look at the unit economics and after a while, you figure out that it is hard to be stand alone. If that’s the case, then finding the right home for it would be great. We were investing in musical.ly, helped musical.ly figure out a way to scale and then encourage them to – when they want to merge into someone else – find a great home in ByteDance so that works out very well for both sides.

ZARA ZHANG: Let’s move on to the rapid fire questions. Who is the entrepreneur you admire the most and why?

HANS TUNG: Someone who’s older and someone who’s younger.

JUSTIN KAN: Everybody probably says Elon Musk. So I won’t choose Elon Musk. I mean I do admire Elon Musk and I think I admire him because he’s always looked so stressed out and he has so much pain that to keep doing it despite the pain? That, I admire. That is a guy who is not enjoying his life. I respect that because he’s doing it for the mission, for the good of humanity even at the sacrifice of the self and that is truly incredible.

HANS TUNG: He might not view it as a sacrifice.

JUSTIN KAN: Maybe that’s right. I hope that’s the case. But the person I would say actually if I can’t choose Elon would be Brian Chesky. Brian is someone I’ve known their entire journey through Airbnb. They used to come by our office across the street in the early days because we were one step and then they rapidly eclipsed Justin.tv and then Twitch. I admire Brian because he always is learning. He’s always trying to level up from somebody who’s a couple of steps ahead and finding mentors. That has inspired me here at Atrium. I’m always like, ‘How do I level myself up?’.

The reason I started Atrium is that I want to build a company bigger than Twitch which is a very high bar. And in order to do that, I’m going to have to continue to be self-aware and level myself up and figure out what I am good at and what I’m not good at and work on improving my strengths and maybe complementing my weaknesses. Really a lot of that’s inspired by Brian. I have a lot of respect for what he’s done to build his company.

There are two younger founders I respect immensely. There’s an obvious one and a non-obvious one. I’ll just go with the obvious one. First, Patrick Collison from Stripe. He’s such a genuine, good person. Who can’t respect that? He’s also very curious. I think his curiosity drives him to really build a company that embodies that. I think with Stripe, there’s so much on the technical excellence side and really investigating the technology side that you can kind of just tell that that’s in the spirit of the company and obviously that has driven massive success.

Maybe the not obvious side, someone I really admire is Daniel Gross the co-founder of a company called Cue which sold to Apple 4-5 years ago. He was very young when I met him. He was maybe 18 years old or something like that. I was just like, ‘This guy is one of the smartest people I’ve ever met.’ He’s just so smart and not just smart on the side of intelligence and technology but he’s also very connected in Silicon Valley, has high emotional intelligence, has been able to like build an amazing network here in Silicon Valley. Now he’s a partner at Y Combinator, is running this AI grant program as well. I always feel like when I talk to Daniel, I’ll learn something.

It’s funny because I would go to him and I’ll go to him and I do for Atrium and I ask him for feedback and advice on my business ideas and plan. And there’s not that many people in Silicon Valley including many of investors who I’m like, ‘Actually that’s the guy I want to talk to.’ And he how much younger than me? He’s probably 8 years younger than me. So that’s amazing to me.

ZARA ZHANG: What’s something you read recently that taught you a lot?

JUSTIN KAN: Something I’ve read recently that I think was really great was this book called The Elephant in the Brain by this guy Robyn Hansen who’s an economist and another guy from Silicon Valley, a programmer. This book is all about the hidden motivations in human behavior that we don’t actually want to talk about.

One of the things that I remember from the book was that brand advertising works not just because it’s brand advertising and you see it so you get that exposure, but it’s also because when you know other people have seen it. So BMW advertising as a luxury brand works not just because you see BMW and you see this cool ad and you think “BMW is cool”, but because you know other people have seen that.

So you know that BMW has a luxury place in their minds. And so that means that basically if you buy BMW then you get that value of knowing that all these other people think it’s a luxury. So I thought that was an interesting observation and it was something that I haven’t thought about. The book is filled with examples of that. And so that was pretty fascinating.

ZARA ZHANG: What do you do for fun?

JUSTIN KAN: Great question. I have a place up in Sea Ranch, California on the California coast and go up there and I ride my ATVs around and throw axes and go on hikes outdoors.

ZARA ZHANG: All right thank you so much for your time. This was fun.

JUSTIN KAN: All right thanks for having me.

HANS TUNG: Thanks for listening to this episode of 996. By the way, we also produce a weekly e-mail newsletter in English also called 996 which has a roundup of the week’s most important happenings in tech in China. Subscribers have told us it is informative and fun to read. The newsletter also features original content and analysis from Zara and me. Subscribe at 996.ggvc.com.

ZARA ZHANG: GGV Capital is a multi-stage venture capital firm based in Silicon Valley, Shanghai and Beijing. We have been partnering with leading technology entrepreneurs for the past 18 years from Seed to pre-IPO. With $3.8 billion dollars in capital under management across 8 funds, GGV invests in globally minded entrepreneurs in consumer internet, e-commerce, frontier tech and enterprise.

GGV has invested in over 280 companies with 29 IPOs in 22 unicorns. Portfolio companies include Airbnb, Alibaba, Ctrip, Didi Chuxing, DOMO, HashiCorp, Hellobike, Houzz, Keep, Slack, Square, Toutiao, Wish, Xiaohongshu, YY and others. Find out more at ggvc.com.

We also highly recommend joining our listeners’ WeChat group and Slack channel where we regularly share insights, events and job opportunities related to tech in China.

Join these groups at 996.GGVC.com/community.

HANS TUNG: If you have any feedback on this podcast or would like to recommend a guest, please email us at 996@ggvc.com.