There is a secret to the products that have been built out of Silicon Valley, which is that many of the largest products that have ever been built, whether these are social media apps, whether these are marketplace companies, whether these are collaboration tools like Slack and Zoom and Dropbox and AirTable and Ocean and so on. What Network Effects tells you is for this style of products, these are products where the more users that use them, the more valuable products become. Andrew Chen, welcome to the show. Thank you for having me.
Man, we were just talking, how did you find the time to write a 400-page book with all of the other stuff that you do? Well, the fun part about it was it was just like having two jobs at once. And this was also one where having the COVID break actually made it so that when you're stuck at home, it's like having your own Walden Pond. You're just stuck in your office and you're like, what am I going to do?
I'm just going to write this whole thing. But no, actually, I ended up doing a bunch of really funny things just to force myself to write. And so I not only put everything in my calendar, I actually turned on all the, I had a separate computer with just the apps for writing on it. And I turned on all the, like, kid protection-safe things.
So I blocked Twitter and Reddit and all my favorite websites from the thing. And then I just tried to write as much as I could. So anyway, it took three years, but now I'm here, which is great. Big lift, man.
It's so funny now that our habits on particular machines mean that we need to create our own, like you say, Waldorf Gardens, that are these little oases of work. And these stupid games you've got a time box which I use as well to lock your phone away. And you've got one of those. Yeah, man.
How important, obviously, so your job advising companies, investing is an executive role. Like, by definition, it's an advisory role, but you do a lot of writing. So how important has having a public facing communication channel being for someone whose main job is kind of a bit more back of house? Yeah.
Well, I think one of the big things that the people think about as an investor, if you break down the skill set, you basically can say there is a sourcing part of the job, right? That's getting, that's making sure that all the most interesting startups come to you and you're meeting them in the first place. There is a picking part of the job where you try to make sure that you're picking the write startups. And that is very, very hard.
It's very random, especially because you're often, in the case of a clubhouse, for example, I met the team when they were two people. They had 500 daily active users when we let the investment. I was one of the first hundred or so users on the product. And so it's so early.
It's so random. So how can you make sure that you're picking the best ones? There's a third dimension, which is winning, which is making sure that the startups that have a lot of options for investors, that they pick you over, over others. And then there's operating and making sure that you're actually helping the companies after the investment.
And so the nice thing about writing a book is it actually touches a lot of different aspects of those four skills. And so when you write a book, it's obviously number one, kind of an advertisement of your skills and your expertise out in the world. That's very helpful in people know me, partly from my blog and from my social media already. And so the book is kind of an extension of that.
Writing the book forces you to refine your thinking down on a piece of paper that you're trying to describe to people. And so that's actually been really helpful for a picking standpoint. Because now I'm like, OK, yeah, what retention rates do I think about? How do I evaluate if a company actually has momentum or not?
And then for winning, by being an expert in this space, it's great. And then operating, it's going to be very, very helpful once the book is out in six days to be able to actually hand the book to entrepreneurs and say, hey, this is how I think about things. And have a high density way to convey a lot of information. Is there a Matthew principle going on here with the best known investors then?
Surely the startups that think that they've got the best opportunity of succeeding want to be attached to the investors and the advisors that are the highest profile because they think this person's going to make me be more successful. So if you've got this very well read blog, this Twitter, this book that's going to be hopefully successful, that means more people come to you. Are you starting to see or has it already happened that it's stratified out into the habs and the have not so little investing world? It's just changing so much.
OK, so there's a great book actually I'd love to recommend, which is called Valley Boy. And it's by Tom Perkins who started Kleiner Perkins, one of the oldest venture capital firms. And if you go back to that period of time, he was literally driving around the Midwest, knocking on the doors of insurance companies, trying to get them to tobacco venture capital. And so because of that whole early period of venture capital, which was in the 1960s, 1970s, was very much about how do we even raise money from pension funds and insurance companies and things like that to even go invest.
And so what that's tended to select for in many of the early years is even if the initial VCs were operators and folks who'd really built companies by the second or third or fourth generation, they tended to actually get people that were more like kind of finance backgrounds. And so we've gone through this long evolution and this circle back where now I think social media and Twitter is just such a huge part of being an investor because now capital is plentiful. And so folks that are amazing on Twitter, like my friend Turner Novak, who only does this post memes on Twitter or you have Ryan Hoover and you have a bunch of these folks that just have amazing social media followers are able to raise money and are able to deploy. So I always joke that a lot of what we've done over the last 10 years of startups has been to give anyone credible two or $3 million to pursue their dream, which I think is amazing.
It's great for the world. But now what we're going to do is we're going to run another experiment, which is to give all the social media influencers and give anyone who's all the CEOs that have a lot of influencers, five or $10 million to invest in people they think are smart and we're going to see if that creates even more startups. And on one hand, you could say, oh, is that, is that, you know, is this good use of money? I mean, but look, I mean startups are really the core source of innovation in the economy and I for one am excited to let anyone pursue their dreams on this, even if I think they're ridiculous because sometimes they turn into amazing rocket companies, electric car companies and things like that.
And I certainly invest in many things that are for kids and teenagers that nobody understands either. So I think it's a great thing. Given the entire lifespan of startups and investing, are you happy that you're here right now? Would you have found it really cool to have been driving around in the 70s or do you think that in 30 or 40 or 50 years time is going to be something incredibly interesting?
Or is this a hockey stick sort of inflection moment? That's a pretty cool time to be an investor? I think it's an awesome time to be an investor, but I always loved the idea of being able to hit the Fast-forward button. If I can, I think of anything, what we're going to, what we're seeing is a rapid decentralization and removal of gatekeepers in the startup industry overall.
And going back to that original example, in the 70s or 80s, it was a very, very small group of people who are investing. It was one street, Sandhill Road, right next to Stanford University, and it was literally operators and former professors investing in the top grad students at Stanford. It was a very narrow group that you're talking about. And through that narrow group, we got great companies like Cisco, like Oracle, like Google, like many of these.
But I think what we're seeing now is just this incredible decentralization, obviously driven a lot actually by the pandemic, driven a lot by the prevalence of remote work, driven by Web3 and the decentralized community. And so I think more and more we're going to see, we're going to think about this idea of Sandhill Road as almost being kind of, and Silicon Valley is no longer exist. I mean, all the companies, all the really interesting companies for the last 10 years have been formed in San Francisco, not Silicon Valley, anyway, not next to Stanford, all everything in the city, all the Airbnb, all the Ubers, all the slacks and so on have all been San Francisco. So I think a lot of this is now, I think, slowly translating into more of like a state of mind.
It's the Silicon Valley state of mind, as opposed to thinking about it as if it's the fixed thing. So I'm very excited about the trend and I think that we're going to just continue. And I think that it's the same thing that's happening in entertainment. It's the same thing that's happening in entertainment where if we're having blogging software and social media means that you to put something out, you don't need to talk to a media corporation, you don't have to get a book deal.
It's the same thing that's happening in music. It's the same thing that's happening in many, many industries throughout. And so it's a really exciting trend. It's kind of dumb that people were still constrained geographically in the age of the internet and it took a global pandemic to liberate the trains a little bit.
It feels like it was probably a little bit overdue. Way overdue. And I think a lot of the hubs that it formed, especially in Europe, in the UK, I was just in London a couple months ago and there could not have been a stronger moment for the startup community there with the number of VCs and investments that are happening there. There's a lot happening in Latin America right now.
There's a lot happening in Asia, Southeast Asia for a long time and obviously in China. And I think what a lot of it, the reason why a lot of it this existed was when I moved to the Bay Area in 2007, if you wanted to learn about these kind of really interesting esoteric topics, if you wanted to learn about, okay, how do I build a product that takes advantage of the pandemic? How do I measure virality? How do I measure a variety of things?
How do I measure retention? How do I measure retention? Oh, there's these things called co-workers. You had to literally talk to the people.
Like you literally had to go talk to the operators that were building the companies and ask them. I originally met, for example, Eric Reese who wrote the Lean Startup many years ago and that company was doing a lot around retention curves. And nobody had written it down. Nobody had written it down.
So I literally just talked to Eric and just asked them how they were doing things and we just talked about it and how he was building the idea around Lean Startup. And I thought it was really interesting. But these days it's like that's all being pushed out into the world. We have endless podcasts.
We have social media. We have books. We have all these things. And I think what that means is that's enabling a much, much broader set of individuals from any country.
And I think it's fantastic. It's a really amazing time to be a founder. What do you think most people don't understand about network effects? Well, one of the case studies that I use in the book is Google Plus versus Facebook.
And I use this story because it's such a fascinating one. We see it all the time, which is a bigger company sees a startup being successful and tries to add a bunch of features and tries to copy a clone it, trying to make it happen. And we're seeing it right now with Twitter and Spotify and Clubhouse for example this year. And we're also seeing it with, I predict that we will start to see a lot of cloning behavior happening for larger companies who want to get it.
Sorry, Spotify creating an equivalent of spaces in Clubhouse? They have built one. What the fuck has it been? It's fine.
I love those guys. They should take their shot. But what you see in a lot of these is, and I use Google Plus as an example, is that when Google saw that Facebook was so successful, they also started a social network project to take it over. And the way they did it was they quickly built a ton of features.
They made it a huge priority inside of the executives. And in order to get growth, what they did was they just put the Google Plus link on the Google.com homepage. You could put any link on the Google.com homepage and would have 100 million users immediately. And that's exactly what happened.
If you go back and read the news reports, what happened was Google Plus had 20 million users and then 50 million users and then 50 million users. And then 50 million users and 100 million users. It looked like it was going to work. And then within two years it was like over.
They like shut it all down. Right? Even though it was successful. And I think the reason is because when you get into network effects, I think this is maybe a good time for us to cover the definition a little bit.
There is a secret to the products that have been built out of Silicon Valley, which is that many of the largest products that have ever been built. Whether these are social media apps, whether these are marketplace companies, whether these are collaboration tools, like Slack and Zoom and Dropbox and AirTable and Ocean and so on. All of these pieces of software ultimately connect different people together for an activity. And so Airbnb is connecting hosts and guests for travel based activity, cloud houses covering listeners and content creators for social media activity and so on.
And what network effects tells you is for this style of product, these are products where the more users that use them, the more valuable the products become. Right? And the telephone is a good example of that. The telephone by itself is useless.
And there is an amazing quote by Theodore Vail, who was chairman of the American Telephone and Telegraph Company, AT&T. That basically says, look, telephone is worthless. Its value completely depends on the number of connections that the network allows you to have. And I think that's true.
And so what that means is, on one hand, that means that if you have a product that has a lot of connections, has a lot of users, it's very powerful. It can grow on its own. It can tap into viral growth. It can use its network to acquire more customers.
It will increase its retention and engagement over time. It will become a better business model over time. These more people will upgrade when their friends are using the product. But simultaneously, and I'll get back to Google Plus now, it also means that if you use a product and none of your friends are on it or they're only likely engaged, then the product is just not valuable.
It's not interesting to you. And that is the cold start problem. That is a cold start problem because a product that is more valuable when more people use it, the converse of that means that it is not valuable when no one's using it. And so the funny thing about Google Plus to go back to this example is when you're inside of a big company, and thank God people think this way, when you are inside of a big company, you want the biggest numbers as fast as you can.
You want to do a big lunch, you want to do the Steve Jobs Turtle Neck thing and get up on stage and announce this amazing new thing that you've done. But what it means is you often just get a spray random spray of users that aren't really densely connected with each other versus I think that this is the whole theory around the cold start problem, the book lays out why it is that so many of these products often start in this small niche and tend to grow from there. That's why so many products start from high schools and colleges like Snapchat and Facebook and Tinder all started in colleges and high schools and growth in there. A lot of the new BDP products we see grow from individual teams inside of a company before growing and taking over the company like a Slack or Zoom or Dropbox, that's how they grow.
And marketplaces like Airbnb start out in Austin, Texas, South by Southwest, a particular moment in time. And Uber starts out in San Francisco and then grows city to city. I think it's really the fundamental explanation for why these products tend to grow in this way. Does this mean that some growth marketers and companies are perhaps not realizing the value of small individual types of interactions when they're first starting out because they're potentially not going to scale over a longer period of time?
That everyone's looking at where can I apply the maximum leverage when can I look at scaling? But I imagine that getting a product from 0 to 10,000 users or 0 to 100,000 users is very different to getting a product from 1 million to 100 million users. That's right. That's right.
Exactly. And I think you have to divide it into these individual phases. And Paul Graham, who is co-founder of Y Combinator, had an amazing essay many years back called Do Things That Don't Scale. And I want to read just a sentence or two from this essay because I think it's so fascinating.
And he's espousing the merits of just doing these wildly unsalable things that don't just don't sound like they make sense. And this is the first one that he talks about. The most common unsalable thing founders have to do at the start is to recruit users manually. Nearly all startups have to.
You can't wait for users to come to you. You have to go out and get them. And he goes on to talk about Stripe and how they literally would individually try to recruit people in the Y Combinator class in order to use the product. And I think it's a, when you're at a larger company or when you come from a mind of broad based marketing, you think about reach, you think about breadth, you think about building an impact across a huge amount of air space.
And the problem is that's just not how startups should create network effects based companies. You have to be very, very manual. You have to start and build very stable what I call atomic networks. These are networks that are stable on their own.
You have to know how many users you need to get an atomic network going. So specifically a product like Zoom, if one of the guys I interviewed for the book is Eric Wan who's CEO and co-founder Zoom. And I asked them how many users you need for Zoom to be valuable for people to use it over and over and over again. And I think we all know that it's valuable with even two or three people, right?
You can have calls and it's great. You can have meetings. It's fantastic. Something like Slack on the other hand is better when it is being used by five or ten people on a team together.
That makes more sense. But if you zoom out something like Airbnb or something like Uber, you talk to the early Airbnb people, they say you need at least 300 listings in a city before Airbnb is useful for that city. For Uber, you want to be able to hit a button and get a car in under 15 minutes. And what that means is you probably need a couple dozen drivers online at any given time in order for it to work.
And so that concept of an atomic network, what is the smallest network that you can build that's stable and can grow on its own and people can use the product successfully, is a really important concept. Because if you can build one atomic network and you can build a second atomic network and a third atomic network, well, you can probably build 10 or 20 or 30 or 50. That doesn't mean that you can build a thousand. That doesn't mean that you can build a hundred thousand of these networks.
Because once you get to that, you start to need to not do things manually and you need to start thinking about scale. And that's when growth marketing becomes really important. That's when thinking about referral programs. That's when you start hiring many, many thousands of people to build these companies.
But here's the funny part, Chris, which is that in a larger company, what ends up happening is it's almost like the companies become so successful that they forget how to solve the cost problem in the first place. This is exactly what I had in my head. So when you've got someone like Google who decides to go into launch a social media platform, the presumption is we're basically too big to fail. We've got all of these network effects already in place because we have access, passive access to, I don't know how much, but an absolute terrifying amount of traffic that you can just put onto the Google homepage.
But you forget that these atomic networks need to be in place. That also it's not just about adoption. It's about use. It's about ensuring that use is congealed around particular groups of people.
It's pointless having a hundred million people all disparate and not connected as opposed to if you had one million people split up into thousands person friend groups. That's right. And so thank God these big companies think this way. Otherwise, my business of startups and but not would not exist.
And then just to add to that, I think the other asymmetry when you're a big company trying to build a new product is that ultimately, every one of these networks, once they get too big, they start to face really, really severe problems that are just hard to solve. One of the big problems is how do you deal with market saturation? You've been tripling, quadrupling, and quin-tupling for so many years that eventually you just run out of people. If you have a product that's being used by a billion users, how do you double or triple or quadruple that at some point?
It's an example of a... Well, obviously all the Facebook platforms are like this. That's something like two billion daily active users. And so, what do you do?
How do you grow that? And you can grow engagement, but it becomes much harder to grow your total user base. Population encouragement, that's what we need to bring. Yeah, that's right.
That's probably what they need. And which is why they're starting a dating app, actually. That's maybe the secret plan in there. And so I think when you look at that, that becomes one issue.
Another issue is overcrowding. Every single one of these products, when you email is a great example, email is amazing. When you're just at a smaller team and you're just emailing a couple people, it's great. As soon as you're somewhere where many of your listeners will have a personal experience with this, if you work at a company with 10,000 people or 50,000 people, your corporate email is like a disaster.
Like, there's just too much. And that's true for social media, and that's true for any one of these. And I think that's why for many of these products, what happens is as they build their networks to become much, much larger, they actually start to hit the ceiling that really limits their ability to grow beyond that point. And I think what it means is that these larger companies that we think of as invincible, we often think of a product like Facebook to be invincible.
Inevitably, what happens is actually you feel like when you use the main Facebook app, because I don't know, last time you went on there, it's mostly dog pictures and birthday parties at this point. It's almost like there's actually just too many people on the Facebook platform. And so you actually want to segment it down so that you can just talk to your friends, right? And no one wants to use the same social network as their data.
And so you get to this very interesting point where these companies that we think of as invincible many times actually are very weak at their core. And that's what provides the opportunity for startups to go zoom in to one piece of functionality in these companies and just pick it off and just do that one part really, really well and build their own atomic networks and scale their own atomic networks to just make that happen. And so, and so I think that's another important conclusion out of the theories of the book. It's a weird situation to get into where for almost all of you in history, we've wanted more information.
So a scarcity of information has been the limiting factor for acquiring wisdom, wealth, health, happiness, partners, whatever it is that you want. And within the last decade, that is flipped from it because it's a big necessity to it being an abundance. And now the main skill that you need is no longer being able to forage for information. It's being able to filter information.
That's that blows my mind. Ten, I was just going to say I went to the Getty Museum in LA and one of the things that they had that I loved was these, you know, the manuscripts that the monks would copy directly. And that's how you manually would make these books. And these books, if you go and you look at look them up, I think it's something crazy like each book was a luxury item because it was like $50,000 a modern day in history.
So if you had like a library, you were just like loaded because you had like a hundred bucks. And that was like, oh man, you're like your 70 lifetimes of monk work that's gone into your library. Yeah, exactly. That's right.
That's right. Now we're now we're now we're now we're inverted. Now it's like we have so much, you know, but that's exactly what creates the opportunity. That's why you see social apps.
Like if you told me that like TikTok could build a social platform and all it was was just going to be dance videos and they were going to be able to go from that to, you know, all these other things. I think a lot of people would say, well, why doesn't why doesn't YouTube just own this? You know, it's their videos, right? But it turns out that actually these niches, if you can build the proper networks around them, if you invent a new media format, it becomes really, really powerful.
That's a really, really powerful. That's a really interesting point. So it's not necessarily about having the widest access to potential audience or the widest access to the market. It's about having a particular type of person on the same platform that creates a magnified network effect.
So you mentioned it before that Facebook now just feels like the most boomery platform ever and it's I don't I use it for work. That's it. It's just to keep up with people and you think, why is that happened? Well, because of how widely it's been adopted, some of the reasons that made it cool in the first place have been selected out.
And now you start to go forward into a more selective sort of social media, something like Twitter, where you don't have to be on Facebook. You're expected to be friends with the people that you are friends with. I don't follow my mum on my dad on Twitter. I'm friends with them on Facebook.
I don't follow them on Twitter. So, you know, I don't that's sorry dad. And the same thing kind of goes for each social media network and each individual product. But it's so interesting to think about the fact that someone may look at let's say someone looks at TikTok, right?
And they said, that's cringe. What are these people doing these dance videos? It's like, it's not for you. That platform isn't for you.
And it doesn't need to be for you. So this presumption that platforms should aim to grow at any cost to just acquire users doesn't really make sense. It's acquire the right users as well as you can. That's right.
And I think, you know, the funny thing is maybe it's just the natural life cycle of these things that they all become boomergy over time. And when they're saying it becomes millenially, I'm going to feel I'm going to want to throw up in my own mouth. Though I'm still in the beautiful grace period where millenially isn't too bad, but and boomeries like a slight when that pivots God, I'm going to hate myself. I'm going to look back on this conversation and go, you self-righteous wanker.
What we doing. Yeah, as a fellow elder millennial, I feel you on this. Yeah, I think that the one of the concepts that exists in all of these networks is that there's often a hard side of the network and an easy side of the network. And what I mean by that is if you take a network like Uber, we would pay $500 to get a driver to sign up on the platform.
But we maybe would only pay $10 or $20 to get a rider to download the app and try it. And the reason is because the drivers just do more work. They're just more important. They actually do.
They actually do a lot. And this concept is true no matter what product category you're looking at within these network products. Content creators on YouTube are just more valuable than an individual viewer. That's why all these companies are paying millions of dollars to content creators.
And that's all happening right now. And because the content creators are the hard side of the network. Funny enough, you can talk to the Tinder folks online dating the attractive members of an online dating platform. They're the hard side of the network.
They're really hard to get. They're really hard to retain. Everybody wants them. And so because of that, I think to go back to the point that you're making, Chris, when you build a new product, you need to have some new innovation that makes the hard side of the network even more.
It needs to be really compelling for them. And what that means is if you're building Instagram, you need to make it so that the photo filters make the content creators on there, make the photos look amazing. Tinder, I actually have an amazing story in the book about Tinder, which is I sat down with Sean Rad, who was an early co-founder and CEO of the company for many years. And shout out to Sean, who also, by the way, had a major court win today over.
It's very contested early history of the company, which we can talk about briefly. But he talked about starting Tinder in the very, very early days that before Tinder, it used to be, it was more like a classified listening, what MASH.COM was set up as. So basically you'd have Bob, Bob would put his profile up on MASH.COM, and anyone could MASH.COM. And the problem is, let's say that Bob is just like a very attractive person, very well-educated, very successful.
The experience for him actually sucks because the way that Sean described it to me, he said, look, you go to work all day and you answer email. And the last thing you want to do is open up your dating app, and it's just full of messages for you to go and answer. That's why you're not here. More email at home.
Yeah. Email at home. Email at home. Email at email.
Exactly. That's right. At least, you know, when you're at a bar or something, you can be like, oh, yeah, these two people are talking to this attractive person. I'm just going to like talk to their friend over on the side or something.
Like in the internet, there was no, in early on, like, getting there's no such thing as people's inboxes or just be flooded. And so what Tinder did, that was, I think, very, very subtle. It's not something that is very well-understood, I think, is that by creating a right swipe in a left swipe, it allowed the most attractive members of the network to control the number of matches that they were getting at any given time. So they were never overwhelmed.
And so they made these decisions like connecting your Facebook and doing all these other things to make the experience for the attractive desirable part of the network more effective. And so that's all to say that I think whenever you're building a new platform for any of these, where it's a new network, you need to have some kind of an innovation like that. Otherwise, what ends up happening is they would just rather use the other thing, the old thing, right? Because if you build a YouTube and then you build something that looks just like YouTube, then it's like, well, YouTube has all the audience and has all the other creators on there.
I'm just going to go use that. You have to do something different enough. And I think this was one of the reasons why I was so excited about investing in Clubhouse is that for folks like you that have been able to build an audience primarily using audio, as you know, creating really high quality content in audio is just different than creating high quality video content, which is different than writing high quality content. And there's got to be a place where people who are really successful at audio to build an audience.
And so I always think about that and try to reset the order of things. Let me stress the clubhouse idea a little bit then. What is it that is helping the hard side of the problem in Clubhouse? Because presumably there are low value conversations where 16 people just shouted each other.
There are high value conversations where you accidentally get Elon Musk, Naval and Eric Weinstein talking about the Kanye West's new album. And that's a people share it and pirate it onto YouTube, like some old school radio station. How do you on Clubhouse incentivize the high quality conversations? I think that Eugene Way has an amazing essay and he was an unbelievable writer.
Such an amazing writer. And this is his kind of bread and butter area. And he has this concept that I love, which is called old money, which is if you are a video creator, one of the reasons, even if you produce high quality content that you may not be that excited about getting on YouTube now, is that there's too much old money on YouTube. People with millions and millions of followers and how are you going to compete if you're going to do something from scratch?
And so in many ways you're incentivized to actually find a new platform, a new network to join, that is just getting started. And if you're just getting started, maybe you have a better chance to get to millions of subscribers. So every new network has this inherent attraction to it that if it works, you can reset the order, you can reset the hierarchy of success and you can climb to the top more easily. So it's a creative opportunity for first move or advantage as well?
Correct. Exactly. And so I think what's happening is in the podcast world, which Chris, Chris, you know so much about one of the reasons I've been reluctant to actually start a podcast has been, I'm just like, oh man, there's so many podcasts out there's so many podcasts out there. It's so hard to build an audience.
Who the fuck am I up against Tim Ferriss? Yeah, exactly. Yeah, exactly. And then you have like, and then the tools are so clunky and the whole thing is just annoying and what should I be doing instead.
And so I think one of the most interesting things on Clubhouse is that most of the content creators on Clubhouse, most of the people that start rooms, most of the people that are creating recurring shows are not podcasters for the most part. They are brand new content creators that are not interested in spinning up the whole stack to build a podcast. They have no interest in trying to build a podcast audience from zero against all the other people. And instead they're joining a new hierarchy where they can they can just build a new show and and if Clubhouse makes the tools very easy if Clubhouse is able to continue growing their audience as they have, then they can build a new network and go from there.
It seems to me like the trend is going from more sophisticated to easier and more frictionless. So if you were to take YouTube and compare that with TikTok. So TikTok has inbuilt editing. It's got music there ready to go.
You can everybody expects you to do with your iPhone. In fact, if you do it with something that's more sophisticated than an iPhone, you almost look a little bit like you're trying too hard. You compare Clubhouse with podcasts. You've got this disgusting RSS feed backend thing with how it distributes.
It's archaic and insane and medieval and primitive. And then you've got press a button on your phone and it streams live to the entire world and people can tune in and watch it. Do you see that trend can't continue forever or else you end up oversimplifying a product out of existence and you just end up with nothing? Where do you see in terms of trends that moving next?
I think just to zoom out kind of historically, right? You have to ask yourself, why did this focus on quality and curation? Where does it come from in the first place? And this is true.
We were talking about books. This is true for radio stations. This is true for starting up a TV show. These are all these things were all very, very hard.
And the reason is because in the pre-digital world, it was all about finite self-shelf space. It was all about having finite numbers of channels, finite number of time slots. And so if you are going to have a retail store that has a limited amount of square footage and a limited amount of shelf space, every single book should be the best book. And so what that does is that affects everything down the chain.
It means that then the publisher wants to select for only the most credential people. They want to select for only the topics that they think are going to sell the best. And then as an author, you end up going in the same thing as well. And again, you can make similar analogies for all the other media types.
And I think the most amazing thing about the internet is there's no shelf space. You can just do whatever you want. And because you can do whatever you want, that means that the very first thing, in my opinion, that gets filled is all the casual content that has never had a place to exist in the world. That vacuum needs to be filled right away.
And so I think that that ease of use, as you say, everything is like you publish a content, you publish text to the internet using, you hit a button, publish text, you hit a button, you publish video, you hit a button, you publish audio and off you go. And that's fantastic. And I hope that continues. On the flip side, though, I also hope that the internet and this, in this ever burgeoning market also provides a ability for the top-end highest quality content to also be successful.
And I think we're seeing it already. We're seeing it because we're in the golden age of dramatic streaming television right now. And the fact that Netflix has infinite shelf space, they can just recommend different things to different people means that they can create these massive budgets to go fund television shows that we've all wanted to watch and make a lot more sense there than streaming. Similarly, I'm a board member of Substack, where they're doing the same.
They have a program called Pro where they've been able to get some of the most amazing authors and writers from all over the place to join Substack. And this new business model of having customers actually pay directly into newsletters to directly support writers has meant that writers who are making, you know, writers are horribly underpaid in the world. And, you know, you get some of the best writers. You can pull them out.
We have writers on Substack that are making 5 million plus per year now. I mean, just amazing. And the reason for that is because the whole market for that is just wildly inefficient. And so I also hope that all of this internet stuff also makes the highest end content, long-form content, serious content, also grow as much as all the short-form silly stuff that we see, all the cat memes, and all the characters.
And all the cat memes and all the dance videos that are out there. Yeah, the two biggest success stories, I'm aware Substack is an absolute monster. But the two that have been closest to my heart or that I've watched them most closely was Matthew Iglesias, so co-founder of Vice. You think Vice is pretty big time.
Had his own TV channel, huge YouTube, huge website, blogging, writing, etc. And he decides to leave to go and do his Substack. And the other one was Scott Alexander, moving from Slate Star Codex to Astral Codex X10. And that is, that blew my mind.
He mentioned Scott's quite open door and he breaks the fourth wall with a lot of this stuff. And he was talking about just how amazing the team had been at Substack, the fact that they'd helped him to port over some disgusting amount of blog posts, like 10,000 articles or something. That was all going to be ported over. And he's now liberated to do his psychiatry stuff as he wants in a new location.
And that was all as a middle finger to the New York Times. It's like, okay, you're going to dock somebody, well, the internet's going to Brazilian Jiu-Jitsu that around, and you're going to end up being the genesis of this person's new life. That's right. That's right.
And so, that's the point that I'm going to talk about. And I'm going to talk about the point about having infinite shell space. Is that it means that as long as there's market demand in these networks that are being formed, all of a sudden you're going to see the hard side of the network emerge and start to supply, whether it's content or it's video or it's, you know, in the case of B2B products, it's meetings and conferences and all of that as well. And so I think that this is one of the really important trends that's happening in the market, which is that rather than thinking about big monolithic networks that are everywhere, everything's open, everything's public, increasingly a lot of this is going towards these private communities, private networks.
Web3 is obviously very interesting trend as well, because what it's going to do is it's going to allow these communities to actually support themselves and to monetize. You look at these projects like Friends with Benefits, you look at these NFTs like, you know, Board Ape since so on. It's very much the idea of network effects baked into crypto and Web3 all together in one. And I think we're going to see a lot of new business models that are going to emerge here.
And it's going to be really interesting. Are you familiar with Shiny Object Social Club? No, I'm not. Tell me about Shiny Object Social Club.
So my buddy is one of the main guys behind it. I've got a couple of friends. Tom is a big part of it. And they have created, it's a Discord server and now a bigger, bigger community.
Connor, my designer, has been a big part of pushing it as well, all around the NFT space. But they've like reversed backwards integrated themselves with creators. They've forward vertically integrated themselves with the people that actually want to buy. They've horizontally integrated themselves into designers, into people that supply, into people that understand coding, into people that have got different NFT projects and different.
And it's just, it's cool, man. You've got a network effect project that has had a community of people create a network around it. That is a social book. It's the most meta thing ever.
But it's sick. He was a question of how to use. We're talking about Web3 and the internet becoming increasingly decentralized. Is that going to change the nature of network effects moving forward?
I struggled so much in writing the book on how much crypto, how much Web3 to put in. Because as you know, it's changing every week. So I'm like, oh man, if I write something and I put it in there, it's going to be outdated in a month. And so what am I going to do?
So I have a smattering of things in there about Bitcoin and Ethereum. But I kind of kept it high level. So we'll have to wait for the revised second edition for me to add all the Web3 examples into all this. But I think that's right.
I think it is maybe the most interesting aspect of writing this book about network effects and seeing Web3 intersect at the same time. Because to go to the example of just Bitcoin at the most basic, people value Bitcoin because other people value Bitcoin. Right? And it's circular in that way.
And in the same way, if you were to start a different type of coin and altcoin, and no one was interested in it, even if you forked the Bitcoin code and you're running all the same code and everything, you could have all the software exactly the same. But if you don't have the network, then it doesn't work. And so at its core, Web3 has this cold start problem in all these different ways and NFTs in the same thing. And crypto gaming is the same thing.
If no one's playing Axie Infinity, it wouldn't be as valuable as what it is. And so I think what we're starting to see is first that all these theories I think are going to apply to Web3, number one. And I think number two, I think what's going to be very interesting is we're just going to see new creative ways that weren't possible in the Web2 era. I'll give an example.
In Uber, one of the programs that I ran at Uber was the Give $10 Get $10 referral program at Uber. And this was an amazing program. We spent $300 million a year on this program when I was there. And this was appearing where we were adding 3% of the world's population to Uber as the app.
It was just growing. It was just talking, sticking. But in the end, we would not be able to give shares of Uber to the drivers or the riders. That's very hard.
You have the SEC of all these things, you have all these regulatory issues and so on. And so it was all about just giving people credit inside of the ride share system. What's amazing about Web3 is that it's really unlocking the ability for the network participants to own a piece of the upside. A lot of Web2 is about maybe you can get a revenue share, maybe you can get some discounts, it's very cash based, but there's no ownership.
And so I think by having ownership, well, what happens? Well, they kind of have a super referral program. You have an amazing referral program because it means that your users actually own a piece of the network and they are heavily incentivized to promote. It's ethical marketing on your service.
That's right. Exactly. And I think that we are just in the very earliest ages. We're going to see very complicated, you know, referral contracts models.
If you refer 1%, this is what you get. If you have 5 people, this is what you get. If you're one of the first 100 people versus the first thousand people, you're going to have various kinds of... And I think we're going to do a wild amount of experimentation and iteration on making this all work.
So I'm very excited from a growth marketing kind of user acquisition lens that we're going to have to try all of this stuff out. Talking about metrics that companies follow, what do you think are some of the most useless metrics that people have focused on? Oh, yeah. Well, I mean, the whole lens of the theory is that the top level numbers are the most meaningless.
Your total number of users, you know, your total amount of revenue. You know, those are what are often referred to as these manatee metrics because they make you feel good. They're the biggest numbers in the whole business. But in the end, the question is, on a given, like, I'm a lot more interested for sub-stack what the top writers are making.
And are they happy? And what's the churn rate on the subscribers on an individual basis? I'm a lot more interested in that than the overall revenue of the business. For a clubhouse, I'm a lot more interested in how many recurring shows are they seeing?
How many content creators are coming back and creating show after show after show? And are they getting the audiences that they want? That's a lot more interesting than the total number of daily active users. And so I think the more operational you are, the more you're focused on that authenticity, the more you need to dig in from the user's perspective and figure out what it is that they're into as opposed to the top level numbers.
I've just realized the original way the clubhouse launched with that text-only invite mechanism, that's a structural way of creating atomic networks. That's right. That's right. Yeah, and maybe I'll talk a little bit about invite mechanics for a little bit.
So what ends up happening that's so interesting with invites as opposed to buying a user off of Facebook or Google? I hate it when startups pay Google or Facebook. If I can avoid it, I try to avoid it. And the reason for that, the reason why inviting is so powerful is because it means that every user that's joining probably already definitely knows at least one person that's already in the network.
Definitely knows that. That's just fact. But very likely knows multiple people on the network. And so what ends up happening is one of the case studies I covered.
I interviewed Rita Hoffman, who was co-founder and originally CEO of LinkedIn. And he talked about how in the very, very earliest days what they found was there's a structure in the professional market, which is that it's very, very hard to get Bill Gates to sign up for a professional network product. Because everyone wants an intro to Bill Gates. He doesn't need to talk to more people, so he's fine.
But there are tons and tons of people that are kind of still operating. They're still in the middle level. They're still very open-minded. They still want to connect with a lot of people.
And they're the entrepreneurs and the founders and operators. And so what LinkedIn did in the earliest years was they basically said, okay, we're going to launch and we're going to give everyone in the company a ton of invites. And we're just going to invite all of our friends and all the people in the tech community. And what they found was that very quickly started to grow rapidly just on its own.
And they eventually made it so that once they got enough users, they could actually remove the invite constraint because just based on word of mouth, on average most users would join in already know a couple people in the platform. But there's an exclusivity element here as well, which adds prestige to being invited. That's right. Yeah.
And obviously, LinkedIn is much more diffuse now. It's much more, you know, it doesn't have that. But like in the earliest years, I think that's right. I think all of these apps, one of the things about invite only is there's a lot of reasons to invite only one is I think the most important and understated reason is exactly the one that you put your finger on Chris, which is making sure that people already have connections.
The second is, honestly, for a lot of these products, they just don't want to scale their infrastructure that fast. If you're growing, you know, at one point, clubhouse is growing 50% a week. It's hard to keep your servers up and running when you're doubling, more than doubling every month. That's very, very hard.
And then the third is you definitely do get a lot of buzz just from people thinking about feeling left out. Now, the funny thing is that's not enough to sustain your product. Because if you get a lot of people coming in and they're all there for the buzz, and it turns out that their friends aren't in the product and it's not working, then it's just not going to work. These days, how many invites do we get into random apps?
It's just not special anymore. And so I think you still need at its core, you still need the engagement and the product to be amazing in order for it to be useful. So, are you with Raya? Yes.
Yes. I've heard of it as Instagram models dating tech millionaires. Is that accurate? I've got it on its own.
No, neither am I. But my buddy is a very famous comedian told me about it. Two Netflix specials, very, very well known. And he said with a lot of a plumb that he wasn't accepted onto it and his invites to the clinic.
But with that, when you sign up, it connects to your phone book and you go through and you see all of your phone contacts that are already users on Raya. And then you select the ones that you think will attest to you being a worthy member. And they receive a notification inside of the app saying, do you know this person? So they're looking for almost like a referral scheme for quality to make sure that people are of the correct caliber.
You had some cool stories about how Tinder started as well. Yeah, that's right. Well, and I just wanted to add on to what you just said, which is one of the fascinating things about viral growth in a case like, I must think of viral growth as you put out a really cool video and everyone shares it and it goes viral. I mean, what is really fascinating is there is a science behind viral growth, which is if you can get a thousand users who use your app to then invite another thousand users.
Well, those users are going to get another thousand users and so on and so forth and it'll just continue to grow. Now, if a thousand of your users only invite 500, then that group will invite 250, right? Because it's about half and then the 251 by 125 and eventually it'll kind of like die down. And it turns out that ratio you can measure as a viral factor and you think you are not.
It's the unknown. Yeah, well, that's right. Yeah, well, now we're all amateur epidemiologists and so we actually know all these terms. Yeah, or not.
Exactly. And so what is happening is if you can calculate that, it means that then you can come up with clever ideas on how it is that you should increase that. And I think this whole references thing, it sounds like a very clever way for them to disguise an invite strategy alongside something that is aligned with their brand and to bring it together. Yeah, and then maybe I'll cover Tinder because it's such a fascinating early story on how they solve the cool problem as our final topic.
So Tinder originally was started Sean Radj and I think you should be on it. And the funny thing is if you think about it that way, that's kind of like an insult. That's like you seem like you're lonely. I think you should probably be on the dating app.
You know, it's not the best sell. It's not the best sell. And so what they realized was they were like, okay, how do we get? How do we get?
We need a couple hundred really desirable people on the app at the same time. And if there's awesome people on the app, people will sick. And so they ended up actually coming up with this amazing idea, which is they were going to sponsor and throw a party for one of the really popular girls on campus. It was a birthday party, but it was going to be amazing.
They're going to like bust people from campus and then they're going to have like rent out this huge house. But they're going to have bouncers in the front. It's going to be very exclusive. And you have to have installed the Tinder app and you have to have set up your profile in order for it to work.
So they got 500 people to this party and they had an amazing party. And the next day, people opened up the app and they're like, wow, here's all these people that I wanted to talk to you that I didn't talk to yesterday. And using that one party at USC, they were able to prove that they could take over an entire campus. And once they knew they could take over an entire campus, then they would try to be able to do this on a second campus and a third campus.
And that became the core of the Tinder strategy. So finding the cool person or the cute girl who had the big birthday party coming up, run it, make sure everyone had Tinder. There's an atomic network for this campus, this campus, this campus. And once you get it up and they start to join, you get USC and you get UCLA, then you get LA.
And then if you get LA, you get New York and you get San Francisco. And it kind of spread from there. But Chris, this is so wonderful to be able to talk to you about the new book. And I really appreciate being on the podcast.
I'm a huge fan and so it's great to be able to talk. My pleasure, man. Why should people go if they want to check out more of your work? So the book, all the preorder links are on coldstart.com.
And then my main blog where I've been writing for over 10 years and I've been publishing, I'm up to almost 1000 essays is Andrew Chen.com. Thanks, Chris. Andrew, thank you. All right.
Thank you very much.