In short
Invest Like the Best - Episode 338: Chris Paik - Venture Investing Frameworks
Episode Overview In this episode of "Invest Like the Best," host Patrick O'Shaughnessy interviews Chris Paik, General Partner at Pace Capital. The discussion focuses on Paik's investing frameworks, the intricacies of venture investing, and various consumer trends.
Key Themes
- Investment Frameworks: Atomic value swaps, business model product fit, and the seven deadly sins of business.
- Consumer Trends: The rise of virtual YouTubers, Apple's Vision Pro, and the evolving landscape of content consumption.
- Philosophical Insights: The history of humanity as a data storage problem, the implications of technology on human interaction, and the nature of innovation.
Detailed Notes
Introduction
- Patrick O'Shaughnessy welcomes Chris Paik, who has a significant background in venture capital, previously working at Thrive Capital.
- The conversation explores frameworks for investing and significant trends in technology and consumer behavior.
The Nature of Investment
- Humanity as Data Storage: Paik discusses the evolution of humanity's methods of data storage and transfer, emphasizing how new computer interfaces affect this dynamic.
- Importance of "Why Now": The need to understand the current technological landscape and its dislocations when making investment decisions.
Investment Strategies
- Assessing Investments: Paik highlights the importance of evaluating investments based on gray areas and the nuances of technology.
- Atomic Value Swap: The fundamental exchange between consumers and businesses; this concept is vital for understanding market dynamics.
- Seven Deadly Sins: A framework for understanding what motivates consumer behavior and how monetization aligns with these motivations.
Emerging Consumer Trends
- VTubing and Content Creation: The rise of virtual YouTubers (VTubers) is transforming content creation, allowing a broader range of people to succeed in media without traditional barriers.
- Impact of Apple's Vision Pro: Discussion on how advancements in technology can enhance human-computer interactions.
Business Model Dynamics
- Business Model Product Fit: Successful ventures must not only find product-market fit but also align their business models accordingly.
- Experiments at Twitch: Paik recounts his experience at Twitch, illustrating the evolution of monetization strategies from advertising to subscriptions.
Challenges and Dark Sides of Tech
- Inequities in Technology: The conversation touches on how new technologies can exacerbate social and economic disparities, citing examples like the impact of Uber on taxi medallion holders.
State of Consumer Investing
- Current consumer investing opportunities are limited, with significant innovations appearing to have plateaued since TikTok's emergence.
- Paik emphasizes the need for investors to engage with new modalities of technology and human behaviors to identify future opportunities.
The Role of Founders and Markets
- Founder's Market Fit: The significance of a founder’s experience and insights related to their market, which can significantly influence a startup's success.
- Sourcing Opportunities: Paik suggests looking at community-driven platforms like Reddit for emerging business ideas and trends.
Key Takeaways
- Frameworks for Success: Understanding the atomic value swap and business model product fit is crucial for venture investment.
- Emerging Content Formats: The rise of VTubers and new content creation tools reflects broader trends in consumer behavior and technology integration.
- The Importance of Human Capital: The U.S.'s advantage in attracting ambitious talent continues to position it as a leader in innovation.
Closing Reflections
- Chris Paik shares personal insights and the pivotal role of conviction in investment decisions, reflecting on a story where he stood firm for an investment in Twitch despite initial pushback.
Call to Action Listeners are encouraged to explore further insights by visiting the Colossus website for additional episodes, transcripts, and resources. They can also sign up for Colossus Weekly for curated content and highlights from the podcast series.
---
This structured summary captures the essence of the podcast episode, highlighting key concepts discussed while making it accessible and easy to navigate.
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Transcript
Automatic transcript. May contain errors.0:00Something I speak about frequently on Invest like the best is the idea of life's work. A more fun way to think about it is that I'm looking for maniacs on a mission. This is the basis for our investment firm, Positive Sum, and it's the reason why I'm so enthusiastic about our presenting sponsor, Ramp. Not only are the founders, Kareem and Eric, life's work level founders, certainly maniacs on a mission. They have created a product that is effectively an unlock for founders and finance team to do more of their life's work by streamlining financial operations, saving everyone their most precious resource, time.
0:30Ramp has built a command and control system for corporate cards and expense management. You can issue cards, manage approvals, make vendor payments of all kinds, and even automate closing your books all in one place. Speaking from my own experience using Ramp for my business, the product is wildly intuitive, simplistic, and makes life so much easier that you'll feel bad for any company who hasn't yet made the switch. The Ramp team is relentless, and the product continues to evolve to save you time that you would never have dreamed of getting back. To me, there is nothing more interesting than technologies that reduce friction for other entrepreneurs to be able to build the thing that they want to.
1:05So much attention has gone to cloud computing, APIs, and other ways of making life easy for founders. What Ramp has done and is doing is build yet another set of tools in this category. To get started, go to ramp.com. Cards issued by Celtic Bank and Sutton Bank, member FDIC. Terms and conditions apply.
1:29Hello and welcome, everyone. I'm Patrick O'Shaughnessy, and this is Invest Like the Best. This show is an open-ended exploration of markets, ideas, stories, and strategies that will help you better invest both your time and your money. Invest Like the Best is part of the Colossus family of podcasts, and you can access all our podcasts, including edited transcripts, show notes, and other resources to keep learning at joincolossus.com. Patrick O'Shaughnessy is the CEO and founding partner of Positive Sum and the CEO of O'Shaughnessy Asset Management. All opinions expressed by Patrick and podcast guests are solely their own opinions and do not reflect the opinion of Positive Sum or O'Shaughnessy Asset Management.
2:10This podcast is for informational purposes only and should not be relied upon as a basis for investment decisions. Clients of Positive Sum or O'Shaughnessy Asset Management may maintain positions in the securities discussed in this podcast. My guest today is Chris Paik, who is a general partner at early stage venture firm Pace Capital. Before Pace, Chris actually spent eight years building and investing alongside last week's guest, Josh Kushner, at his firm Thrive Capital. During that time, he sat on the board of Twitch. Our conversation explores Chris's frameworks for investing, from atomic value swaps to business model product fit and the seven deadly sins.
2:48We also discuss different consumer trends like the rise of virtual YouTubers, Apple's Vision Pro, and why everyone graduates off of YouTube. Please enjoy my conversation with Chris Paik. So Chris, I think an interesting place to start because it's something you said at lunch that I don't think I've seen you write about elsewhere is the history of humanity as a data storage and transfer problem and how that relates to new kinds of computer interfaces. We're lucky that we've now gotten to see the latest Apple Vision Pro, which probably figures into this discussion somewhere. What is this big idea of yours and why is it interesting for those interested in technology?
3:25Let me caveat this with it's probably in the bucket of half baked and still in the realm of theory and the world of audiophiles. There's this term called lossless, where if you are an audiophile buff, you're trying to get as close to the quote unquote source or original thing as possible. And so compression algorithms are your enemy. And it's what drives people to spend insanely ungodly sums of money on their audio setups. And I think this concept of lossless is really interesting as it pertains to data and data transfer and what are high fidelity experiences, what are low fidelity experiences, what are lossless experiences.
4:06If we think about data between humans in the same generation, data across generations, information, the ability for us to transfer information, store it, and access it is so important to societal efficiency. It prevents us from reinventing the wheel time and time again. You can learn from other people's mistakes. And so one thing that's interesting to think about is songs. Songs as a compression algorithm for data. We are wired in a certain way to feel certain tunes being catchy and we can remember song lyrics. But if you think about it, the oldest forms of songs, poems, epics, were meant to transfer data between people and across generations.
4:58It's this data compression algorithm and perhaps the original form of data storage. And when we think about the advance of technology that have allowed us to create these step functions in data transfer and data storage, notably the written word, huge, huge advancement. The biggest change though, happened with the invention of the printing press because it took stored data and made it scalable to distribute. So all of a sudden there wasn't one copy of one thing. There could be an infinite number of copies or a multitude of copies of the same thing. And so if you think about individual conversations as an amount of data that's being transferred, we're having this conversation right now.
5:45If it weren't recorded and it weren't distributed, the amount of surface that that data could impact ends, you and me, in this conversation. Or each of us take a little bit of it away or something. High lossiness. And if you think about the invention of the written word, that was it. Somebody writes it down and then it is limited to that single copy being consumed. The invention of the printing press created this incredible distribution mechanism for that same data to be consumed. You look at the internet as another form of data transfer, data storage. One thing that I think is underexplored is the impact of data transfer speeds as actual why now reasons for companies existing.
6:36Let's look at mobile social networks and the order that they were founded in. So Twitter 2006, Instagram 2010, Snapchat 2011, maybe most recently TikTok, nay musically in 2014. It could not have happened in any other order because of the minimum packet size of the content on that platform. So Twitter was text-based, smallest packet size. It was actually able to run over SMS rails. And then when mobile bandwidth was in its infancy, it was still sufficient to transfer those kilobits of information. But it basically wasn't until 3G until Instagram was possible. You had compressed images. And in order for the average person to have a high fidelity experience, you needed wide distribution and attachment rate of that.
7:32Snapchat coming shortly thereafter, also imagery and short form video. And then TikTok three years later with longer form video with audio also part of the consumption experience. The thing is TikTok couldn't have started between Twitter and Instagram. And interestingly enough, I would argue that Instagram could not have been started after TikTok. With all due respect to a number of great founders building great innovative things. I would argue things like paparazzi, B-Real, because they are not strictly more bandwidth consumptive than TikTok, the why now, the technical why now is not sufficient.
8:16And if they were to have Darwinistically existed and survived, you would have seen them emerge in the Instagram Snapchat era of 3G. I think a lot about data transfer between humans across generations, but then also the read-write speeds of humans to computers. So our eyes read data at about, I think it's 100 megabits per second, basically ethernet speeds. Our whole environment, not just reading text, we read what we can see. Yes, exactly. The amount of data that our eyes ingest is about 100 megabits per second. Compare that to linguists who have studied the bit rate transfer of spoken language is hours of magnitude slower.
9:13Most current common spoken languages converge at a data transfer speed of 39 bits per second. Wow, tiny fraction. There are like a number of things that bound that data transfer speed. It is both limited by the write speed of a human voice and the data processing speed of our audio, of our ears. Interestingly enough, you think about people who can listen to - Podcasts at two times speed. You can't listen at five times though. So actually read speed is twice as fast as the spoken word write speed. Spoken language, the limiting factor is actually human write speed. And if you think about voice as an interactive modality to computers, that also suggests that a voice interface will also be limited in a data input write speed of 39 bits per second.
10:04Some examples may highlight things like imagine a fighter jet that's voice controlled. It doesn't work. It doesn't work for a number of reasons we as humans, we've created tools to advance our haptic ability to input data into computers, whether it's the keyboard or the mouse or the touchscreen. I think there's something interesting to think about. Darwinistically, if voice had actually been meant as a input mechanism between humans and computers, wouldn't we have also come up with some tool to advance our data input via the sounds we can make. We would have come up with some sing-song language, or we would make specific sounds that map to certain data, but we haven't.
10:53And I think that begs the question of maybe voice is just not meant to be input mechanism between humans and computers. Maybe we have a structural advantage with our other senses, with touch and haptics. And that is actually the highest throughput data transfer between humans and computers. And it will be for the foreseeable future. We're not going to see, because sight is our fastest read speed from computers, we're not going to see a more dominant interface where computers transfer information to humans that is not through sight. Theoretically, this would argue that because voice hasn't actually been effective as a human computer interface, it never will be.
11:48There are limits to this framework. And again, I'll sort of caveat it with like it being half-baked. But I think the most interesting part of the Vision Pro is not the screens. It's going to get more intimate. Screens get more intimate. We will always advance along better, higher fidelity, more pixels, you know, screens. But the modality won't change. It will always be screens. It will always be sight. What is changing, though, is the input mechanism from humans to computers. So every time we have advanced in human to computer write speeds, we've seen a complete paradigm shift in what is possible.
12:32So you started with the keyboard and that was fine. It was good. It was limited basically by the number of words per minute that somebody can type. Average person types at like 40 words per minute. But then the mouse was invented. And all of a sudden, there was this watershed moment of advance on how much humans can write to computers as data transfer. I think the thing that happened with the iPhone that people misattribute to like, well, it's like mobile, smartphones, things like that, is actually just multi-touch. Multi-touch was the innovation that made the iPhone in such a paradigm shift of human computer interface.
13:11Because we had plenty of smartphones. You remember T9, your Nokia dumb phone? Think about how ridiculously rate limiting that was as a data input mechanism for humans to computers. So enter multi-touch. Apple bought the company, owned multi-touch, and patents around multi-touch two years before the iPhone was launched. That was a thing that allowed humans to write more information more efficiently to the computers that they were using. So I think the most interesting part about the Vision Pro is it's eye tracking and hand gesture tracking. Because to me, those represent advances in humans being able to write information to computers.
13:57If you think about eye tracking, think about the process of clicking on something with a mouse. You look at the thing, your brain subconsciously is thinking, okay, I need to now physically manipulate my hand to move this hardware mechanism to then move a digital representation of something to then meet where my eyes are looking. And then I click. With eye tracking, you basically get to short circuit that entire lossy physical loop. I feel like that's why you hear from early users, it feels telepathic. It's such a gain of efficiency of human rights speed to the computer. I think the hope is that continued advancement of human rights speeds to computers lead to these watershed moments of advances in human computer interfaces.
14:47I think it's a great excuse to talk a bit more about this notion of why now and how that critical question plays into your view of investing strategy. You gave an example around the social networks. In each case, they were newly possible because of some underlying platform change or technology change. Do you think the answer to the question really needs to be ultra compelling every time you make a technology investment? To say as much as you can about this idea of why now. Conceptually and from a platonic ideal perspective, yes, every venture investment should have an essential, very strong, compelling reason of why now.
15:26Whether or not that is actually perceptible, measurable, and articulable in the moment of having to make an investment decision or deciding to found a company, that's a totally different analysis. But from a philosophical perspective, yes, venture is a pretty unique capital instrument. And if you think about efficient market hypothesis, where if there's an opportunity, people create companies to take advantage of that opportunity. If venture is supposed to be exposed to companies capable of explosive enterprise value creation over very compressed periods of time, those opportunities shouldn't exist unless crazy dislocation has happened in the market.
16:14Because unless some crazy dislocation has happened in the market, there shouldn't be the opportunity to capture such a large integral of value. And so if there is no compelling why now, and maybe it's technological or regulatory, I would make the strong argument that that is not a venture. That's not a suitable candidate for venture capital. When you're looking at an opportunity and you're thinking about that question, how do you go about poking at it? Because right now you might be able to say, okay, a company enabled by the Vision Pro, I get it. Why now is that this thing's new? And for all the reasons you just described, it's exciting.
16:56People get this. People understand that dislocations create opportunity and they flood it with capital. They flooded Web3 with capital. This happens over and over again. This is old, old history. But how do you go about in the gray areas assessing this? Because not every great company was created right on the back end of some major identifiable tech change or something. So what does this feel like in practice? What sorts of questions are you asking of the company of yourself? This seems like an incredibly valuable exercise in the canonical question, is the founder more important? Is the market more important?
17:29I think what you're saying is the market is really, really important and great markets that can create a lot of enterprise value come from dislocations. So I'm just curious how you prosecute this when you're looking at a company. So my best answer to that is it's hard to at a micro basis. I think the way that I choose to spend my time, which is hanging out in the corners of the internet, on the edges of things, that increases the likelihood that things that are happening there are happening because of some new advancement. And so in some ways, if you choose to spend your time in those spaces, you get a higher natural hit rate of the things that you are exposed to being possible because of a why now.
18:16It's a good example. Is V2 being a good example? VTubing is probably a good example. VTubing is - You're going to have to explain what this is. Super weird. So VTubing portmanteau of virtual YouTuber. VTubers have 3D models. They build in Unity or Maya. Those are the digital representations of the content creator. So they use motion tracking hardware and software to capture their movements in their face and their body. And those movements end up puppeteering this avatar, this digital representation of themselves as a digital content creator. So you can watch a stream of a VTuber and you're watching the anthropomorphic fox or what have you.
19:02And there's somebody somewhere sitting with, maybe they're wearing motion capture hardware on their body, or they're using cameras to track their face, and they're creating content as that character. And I think what's so interesting about that is, and franchises a completely new set of potential content creators. For better or for worse, to be a successful content creator in this day and age, Instagram or TikTok, you basically have to be good looking. It's sad, but I don't make the rules. And what this allows is for people who maybe aren't good looking, but are very entertaining. They're super compelling.
19:46They're funny. This allows them to achieve the same level of success when they previously were completely disenfranchised. I think that's so, so, so powerful. I think a lot about jobs that might exist in 10 years that don't exist today and jobs that exist today that didn't exist 10 years ago. Like if you told me as a 13-year-old kid that I could make a million dollars a year streaming video games on the internet, maybe I would have lived a very different life and pursued a very different career. But when I was growing up, that was unheard of. That was literally impossible. No one was doing that.
20:21And now it is eminently possible. People can do that as their careers. And so I think a lot about at a micro level, how people reallocate their own labor hours, where they are advantaged, where they're disadvantaged, where they are newly advantaged. And as labor flows to higher states of entropy, because of these new advances in technology, whoever is building that aqueduct, that water naturally flows down, will greatly benefit. Let me say it back to make sure I captured the idea right. it's incredibly valuable to offer, enfranchisement, I guess is the word, to a new class of creators that previously could not have been productive.
21:07And if you do that, this is like the whole, I think it was Bill Gurley or somebody at Benchmark told me like, you can't start a new social network with existing celebrities. It has to be with new celebrities that you've made possible. Is that a fair summary? I could not agree more. That's right. Charlie D 'Amelio. Yep. TikTok. TikTok. She's a dancer. If you think about it, dance as a form of content disenfranchised on platforms prior. To really appreciate dance, you need to hear the music. But audio was never a native endemic form of content consumption on Instagram. It was always visual or Snapchat, visual.
21:44And TikTok was the first platform that made audio a native part of the content consumption along with video. And so dance was elevated from this completely backwater. Maybe if somebody happened to have their audio on while they're scrolling through Instagram, it became a first class citizen of this new economy of TikTok. And it totally makes sense that then the number one most popular TikToker is a person who has previously disenfranchised that is now currently enfranchised. Can you explain for people, almost nobody listening will know this term, who Code Miko is? Yeah, totally. So Code Miko is a streamer on Twitch.
22:23She is a VTuber. The woman behind it, Yuna, is wildly talented. She 3D animator and was inspired to quit her job and start Code Miko as a Twitch channel because of what was then possible in VTubing. Yuna wears motion capture hardware, basically all of her body. That inputs motion tracking information into an environment that she has created in Unreal Engine that real-time animates a character called CodeMiko in a essentially game environment inside of Unreal. Her stream is basically one camera angle of that game environment inside of Unreal. And CodeMiko is the content creator, basically a player, represented player inside of that environment.
23:15What's crazy to think about though is what happens when CodeMiku is not the only player that's represented in that in-game environment. What happens when someone else who's also wearing motion capture hardware software that is also a VTuber wants to represent themselves in the same three-dimensional space? I think there's this weird corner of the internet in Grand Theft Auto where people create content on roleplay servers. We're playing Grand Theft Auto. I'm roleplaying a cop. You are roleplaying a bank robber. You actually have to rob a bank and I am supposed to chase you down as the cop, pull you over, and apprehend you.
23:56The beauty of this is the game basically turns into an improv stage. And you have this network effect of content creators where if I, as a new content creator, get to drop into this existing role-playing server, it's so much easier for me to create compelling content because there are all these actors that are already creating amazing content. And so it's interesting to think about VTubers who exist as these digital content creators the same way as characters in Grand Theft Auto exist on these role-play servers. and what is possible when they are networked together and exist in the same three-dimensional space.
24:40How do you think about that from the perspective of an investor trying to earn a return? I totally get the idea that a great way to back into great why nows is to hang out at fringes. That's the whole William Gibson idea of distribution of the future being here already. That's my favorite. The future is here, it's just not evenly distributed. Exactly. So I love the strategy to find interesting things. It's a great sourcing strategy where there might be a compelling why now that you'll be able to tell very coherently 10 years from now, but it would be hard to articulate now. This one's weird for most people, probably very weird.
25:12How do you then leap from that to like, oh, investment opportunity. I could invest a million dollars and get a hundred million back or something like that. It seems like a big leap. So I think this is where an economic framework comes into. I think there's this concept of product market fit, which we all know very well. And then this other concept of business model product fit. Maybe you could argue that along the spectrum of 0 to 100 of product market fit, companies, products, arrive at their fullest potential of users using the product. But on the scale of 0 to 100 of business model product fit, the company approaches its fullest potential of enterprise value.
25:56So you can have a product that is amazing product market fit. Open source, for example. Open source is a great example of product market fit with zero business model product fit. Linux, incredible technology. Linux itself, there's no enterprise value inside of Linux. I think this is arguably what's happening with Twitter. Twitter is a great example of a company that has product market fit and is struggling to find business model product fit in this very public stage. So when it comes to going back to your question of how do you evaluate something that has potential product market fit at the edge, maybe a good candidate for YNAO, maybe not a good candidate for YNAO, I do through the lens of business model product fit.
26:34There are only so many business models that exist, arguably. You can buy widgets for X and sell the widgets for more than X. That's a business model. One of my favorite business models is you can help people make money and then you take a cut of the money that they make. I think this applies to most marketplaces. There are other business models of you pay me to save you time, productivity tools or the convenience economy, delivery, things like that. And I think basically every company distills down into one of these core business models. When it comes to something like a CodeMiko, I think a lot about how these emergent VTubers are choosing to spend their time, what must be true for them to allocate their time in this way, and where the business model opportunity might exist.
27:33So if people at a micro level are allocating their labor in a way that they previously didn't or weren't able to, they're likely doing so only because it is economically attractive for them to. They're making more money in a way that they couldn't have done before. And so likely the best business model there is the second one we talked about, help people make money and then just take a cut of the money that you help them make. With Mikoverse specifically, the business almost certainly is going to follow the Chris Dixon's come for the tools, stay for the network. Yuna has created this incredible, incredible streaming studio setup that's optimized for VTubing.
28:17And what we did is we backed her to take that, flatten it down into something that is a single runtime and give it away, democratize the tools. It's very technically hard to become a VTuber if you don't know anything, if you're a non-technical content creator. So we backed her to take her entire setup, her workflows, condense them down to a product that then could be disseminated as this democratizing tool set to empower and enfranchise a whole new set of would-be VTubers. And then the long-term bet is if there is this network effect, the supply side network that I was articulating with the Grand Theft Auto role-playing servers, if there's a supply side network effect of content creation, that is this compounding mode where you get this fundamental network effect that advantages you against would-be competitors that offer similar tooling.
Read the full transcript
29:07And then you can layer on things that allow the people that are using the tool inside the network to make more and more money and take a share of the money that they're making. Specifically, I think one of the insights with CodeMiko and Mikoverse is interactivity. Live content has been searching for a monetization mechanism. I think one of the challenges of existing monetization mechanisms like advertising or subscription is that it doesn't take advantage of the natively new possible things with live content. If we're all to crib the benchmark, any new network or new platform will home grow its celebrities.
29:52I would argue that any new modality of creation will home grow its own business model that's unique to it. For better or for worse, other content areas have long pioneered this monetization insight, notably cam girls. Camgirls on the internet monetize very well through interactivity. You are watching this live stream. You pay some amount of money. You are basically immediately gratified with the output of the intended intention. That is actually what is now natively possible with live content that was not previously possible with video on demand. And so what CodeMiko has done is taken the same principle of interactivity, where if you watch her stream, there is a menu that you can see where you can donate some amount of money and actually manipulate the content you're seeing.
30:56So there are these digitally defined interactivity. So like, I want to drop a piano on this VTuber. I want a car to race across the screen and hit them. and these are all digitally defined, coded inside of the environment, also built in Unreal environment, you get the same instant gratification. And it turns this stream basically into a game that the viewers play by paying. And what's wild is that the streamers are making basically 10 times as much money using these interactivities as they were before. So for example, there's a streamer, awesome, amazing streamer, great VTuber, pre-Meekover software was earning about$2 ,500 a month as a streamer streaming 60, 80 hours a week.
31:50Using the Meekover software with the native interactivity made that in three streams. Wow. Crazy. And so that economic value proposition is this underpinning that will drive adoption of the tool. And then with enough adoption of the tool, the opportunity of the network opens itself up. And then you can imagine this entire ecosystem of, well, maybe there are people out there that are just really good interaction designers, interactivity designers. They're super creative. And maybe somebody is going to make a theme park that these VTubers can digitally visit with their own set of interactivities that the audiences can manipulate.
32:30What happens when audiences want to trigger interactivities against other VTubers that are in the same environment as the VTuber that they're watching? I think this is a super exciting, underexplored frontier of content consumption that is also almost really perfectly aligned with a net new monetization mechanism of interactivity. You've got a bunch of interesting ideas around value creation that I'd love to explore to keep drawing this picture. The first of which, maybe the most essential of which is this idea of the atomic value swap. Can you describe what you mean by that term and some of the implications of it for thinking about companies?
33:08The concept of the atomic value swap is what is actually going on? What is the exchange that is happening? Let's take this to a lay example. I walk into a grocery store. I want to buy a head of lettuce. I pay$2 for this head of lettuce. The value swap is$2 in exchange for the secular to secular exchange of money for goods. That's very easy to understand. If the grocery store wanted to charge$20 for that head of lettuce, that exchange probably wouldn't happen. The imbalance, the asymmetry between what is expected in return for what is being delivered won't achieve a market clearing mechanism. What's interesting is applying this to things that aren't maybe as easily measured from a quantitative perspective.
33:58It boils down to why do people do things, period, because it costs everyone something to do anything. And so what are they getting in exchange for the time expended, the effort expended, the money allocated? And how durable is that exchange? How repeatable is that exchange? Does it increase in attractiveness over time? Does it decay in attractiveness over time? Is it a one-time thing? Well, let's think about, I don't know, a company. Twitter's an interesting one. Yeah, Twitter's a really interesting one. Let's think about why people do things on Twitter. Why does a content creator create content on Twitter?
34:43I would argue that somebody creates content on Twitter to increase the distribution of their thought and future thoughts, future content. No one makes money on Twitter. So people create content on Twitter to become more famous, become more successful, basically build their distribution. And it's baked into the follower model in and of itself. Conversely, there are some platforms that no one makes content for to increase distribution. They only do it to monetize. Patreon is a good example of this. Or Shopify. Nobody creates content on Patreon to increase their distribution. They are likely increasing their distribution in other spheres, whether it's they have a podcast or they have a YouTube channel or a Twitch channel.
35:37They are using Patreon as a monetization mechanism for the pre-existing distribution that they already have. Going back to Twitter's atomic value swap, This is why the idea of a paywall is so scary for Twitter. Because the only reason why people are creating content on Twitter is to increase their distribution. What happens when you introduce a chilling effect to distribution? Arguably, this is why I think Medium failed as a product. Medium originally was a tool that allowed people to more aesthetically or on brand in the way that they wanted to distribute their content. It was good content. Theoretically, it would be more widely disseminated.
36:26But Medium wanted to monetize their traffic and they thought they had enough leverage to install a paywall on the consumption side. The issue with that is that violated the original atomic value swap with content creators originally. When you change the balance of the values on the scale, sometimes it no longer makes the original atomic value swap actually market clear or repeat. And so when Medium introduced this paywall on the consumption side, all of a sudden, every person that was creating content was incrementally disincentivized to create content because the proposition of uncapped distribution no longer existed.
37:08There was a ceiling, there was this exogenous force, this compressive force on what distribution they could attain. And the biggest problem is we don't exist in these platforms in a vacuum. There are always alternatives to these platforms. If I was an aspiring blogger or content writer, and I was getting strictly worse distribution through a platform with a paywall, I'm just going to reallocate my labor and time and energy to a platform that has unbounded distribution. It will be interesting how Twitter navigates through finding business model product fit. It has product market fit. The biggest risk is that their attempts business model product fit violate the original atomic value swap, which gave Twitter product market fit.
38:06And it all starts to inwardly compound on itself negatively. Is there a difference between your notion of atomic value swap and Christensen's idea of jobs to be done? Is the key idea under the atomic value swap, just that you know what it is and that many don't know what it is. And if you do, you can then all of a sudden unlock a lot of potential. Yes, I think that's right. I think that maps very closely, if not, is the same thing. I think every company, every product, generally speaking, has multiple atomic value swaps. You basically have as many atomic value swaps as you have constituents. So as you pointed out, on the consumption side of Twitter, I, as a user of Twitter, am paying with my time in exchange for entertainment.
38:51Twitter, as a company, gets to take the integral of time across all of its user base and then turn around and sell that to advertisers. The advertiser atomic value swap is pretty straightforward. It is a secular to secular value exchange of what's the return on ad spend as an advertiser on Twitter. And if it's there, it happens. If it's not there, it doesn't happen. As a user, same thing. I'm going to spend my time so long as this is the most entertaining way for me to allocate an increment of my time. The interesting thing is neither of those things happen if the original value swap with content creators doesn't happen.
39:29That supply side content creator value swap for any UGC platform is so sacred. Once that's compromised, it's basically the beginning of the end for it. It will just downward spiral and be competed away by other platforms that are willing to reward those incremental content creators with a more attractive distribution value proposition. I'm interested how you think about things like elasticity of demand and sources of value creation overlaid against all the stuff that we've been talking about. I've seen you write about the seven deadly sins before and the degree to which each sin is monetizable.
40:12This probably maps onto your business model to product fit a little bit too. Walk through this idea for us because I do think all of these are devices I think of as ways of evaluating a company or a product or a team even. I love the stable of questions that you've effectively created. So what's the seven deadly sins one, how does it map onto the other concepts? Well, first, I feel like the seven deadly sins have gotten a bad rap. I feel like they should be rebranded as the seven core motivators, whatever. I think the seven deadly sins are basically the essential reasons why humans do anything.
40:48I have a relatively Kantian view of altruism. I think people generally do things because they're self-interested without exception. In terms of elasticity of spend, I don't know if the seven deadly sins themselves have a spectrum of elasticity of spend, but I do think that Maslow's hierarchy of needs does. Maslow's hierarchy of needs, like food, shelter at the bottom, and self-actualization at the top. I think humans' elasticity of spend basically increases as you ascend Maslow's hierarchy of needs. Will I pay$100 for a head of lettuce? No. Will I pay$100 for better education for my child? Absolutely.
41:33And so I think it's interesting to think about where different things land on Maslow's hierarchy of needs when companies offer them. And then what is that elasticity of spend. If we look at gaming, for example, one of the greatest business model innovations in gaming over the last decade, two decades, is the invention of what are called cosmetics. Cosmetics do not change gameplay. They only change how in-game assets are visually represented. Think of it as clothing. You're wearing a polo shirt. I'm wearing a button-down shirt. They serve as functionally the same thing, but they say different things about us.
42:17They're different self-expressive choices. In some ways, they probably encroach along self-actualization of this is who I am. I dress in a way to communicate to other people who I am. It's a nonverbal medium of communication and self-expression. And so cosmetics are this incredible innovation that has occurred in free-to-play gaming. Bucket free-to-play gaming monetization in two categories. There is freemium monetization and then cosmetic monetization. So in the beginning of free-to-play games, the general consensus was, hey, we're going to give this away for free, but then we're going to make it kind of hard to continue playing it unless you pay.
42:59Or you can pay and save time, or you can pay and you get this unique asset in-game that makes it a little bit easier. The issue with that kind of monetization is it leads to what are called pay-to-win dynamics within a game, where it's fine for the beginning part of the game when it still feels meritocratic. But as gamers advance to medium and end game, they start to become increasingly disenfranchised unless they pay. And it starts to feel futile. It's almost like the game of Monopoly. No one actually plays the game of Monopoly to the end of the game. Right. It's foregone conclusion halfway through.
43:41And that's the issue with freemium free-to-play games. The elegance of cosmetic monetized free-to-play games is there is no difference in the game experience that someone gets if they pay or they don't pay. And the axis along which the company monetizes is purely along the self-expressive spectrum. There's only like 10 of these skins out there. And I think it says something about me as an individual if I have one of these 10 skins. There's a reason why Bernard Arnault was the contender for the wealthiest person in the world. Luxury goods, things that we wear, the things that we consume, they are higher up on Maslow's hierarchy of needs, higher elasticity of spend a higher willingness to pay for that social signal.
44:34I would argue that Bernard Arnault has dramatically benefited from Instagram and TikTok as well. The same$3 ,000 handbag that you purchase now is signalable over a dramatically larger surface area than 20 years ago, where it was basically only in person. And that drives this parallel demand for consensus signals. So for every Louis Vuitton, there are a hundred mid-tier luxury brands that no longer exist because society has converged this as the dominant social signal. We really haven't talked about people behind these businesses at all yet, which we're leaving to believe that you are like a market maximalist.
45:11You're thinking more about what is the enabling dislocation or something like you talked about earlier that makes this new idea that I'm being pitched possible? And what is the dynamics of the market as it might grow? and we could talk more about that in the future, but I want to spend a minute on the founder side. What do you care about, look for, watch for, prefer when evaluating the people behind these ideas? Assuming you've gotten there on, okay, there's some market story that is really compelling here. Now I have to focus on the people. What do you care about there? I think there's the traditional things like aptitude, integrity, sociability, that kind of thing.
45:53All those traditional vectors are super, super important. They evaluate the person's ability to execute, recruit, inspire, delete, leadership qualities, future fundraising. One of the things that I deeply care about that maybe others don't is what I would put in the bucket of founder market fit. To use an analogy, in the music industry, there's this term called the sophomore slump, where usually the second album that an artist puts out, it's kind of not good. In many ways, I think about the same journey of a founder. Investors get to cheat. You get to intercept that artist the year before they released their first album, but you weren't there for the 10 years of earned insights that it took them leading up to it.
46:40And maybe more tactically, I might be able to articulate founder market fit and earned insights, how they manifest. So if a founder has the correct intuition about a product, and so you start compounding those things, and you sum total, well, we don't have to test this thing because I intuitively know for X, Y, and Z reasons, I have this earned insight. I was the customer, and so I know that we need to build it in this way. And then the company and the product building process is so much faster because intuitively, the leader knows exactly what needs to be built without having to ask anyone else.
47:21So I like to think about founder market fit, earned insights, how long of a period of time this person has basically been collecting the dust of insight over usually very long periods of time and then are able to compress it into these diamonds that somebody else starting from zero wouldn't be advantaged with. So it's really a pathway of slowly versus quickly. Earning an insight literally could just be that simple. There's just some core thing that you couldn't have known had you not spent the 10 years in the muck, either as the customer or as someone trying to produce the outcome or whatever.
47:59If you think about other sources of where these people might start to emerge. The idea of sourcing from the edges is really interesting to me. I've seen you write that any Reddit sub community that has a million people in it probably could support a venture scale outcome business. What do you think about everything going on with Reddit? And I just would love to spend more time exploring other pockets of sourcing advantage, I guess is what I would call it, that you use or have used in the past. But we'll start with Reddit. It's a an interesting company, I think of Reddit as basically an interest graph.
48:36There are plenty of subreddits that people are subscribed to that aren't suitable for social media. There are divorce subreddits. There are grief subreddits. Nobody's going to post on Facebook or Instagram, hey, I'm going through a breakup. Yeah, this is who I am. So there's this entire side of who people are, what they're interested in that isn't met by other platforms. And when I think about Reddit, it is really an interest graph. And as such, if across humanity, users of Reddit, there is enough of this critical mass of people who express interest and engagement around the same interest, I think that is a good litmus test for it having enough critical mass to birth something capable of significant enterprise value capture.
49:27It's a silly example, but if you follow the WallStreetBet subreddit, where it's basically degenerate gamblers... Robinhood lines up nicely. Exactly. You could argue that they go hand in hand. There are some really interesting subreddits out there. There's a subreddit just dedicated to 3D printing firearms. As you can imagine, wildly popular. I think Reddit is such an underexplored, fertile ground of signal, almost like hesitant to say that it is. Most people aren't curious enough to actually explore all the corners of it and glean it for insights. What do you make of what's going on there now, which is this revolt?
50:07I don't know what else to call it. And maybe this is a great example of the conflict between a business model and the value swap for the customer in this case. But just since it's happening right now and you hang out there, I'm curious what you think about everything that's happening and how you would handicap what might happen in the future. I actually think people's prognostication of what will happen with Reddit is relatively indicative of how much they understand the way that internet platforms work. I think my punchline conclusion is that Reddit mods think they are landlords when they're actually sharecroppers.
50:43For better or for worse, whether it is justified or not justified, mods who contribute their labor for free in exchange for power are doing so building their castles in somebody else's kingdom. That kingdom is Reddit. And so it's kind of implicitly part of the social contract. If you do that, you are at the whim of the governing kingdom. I understand why they're upset. I understand why the protest is occurring. But if I were to predict what will happen, mods who black out their communities, interestingly enough, are probably exactly the mods you don't want on the platform. The ones that are willing to compromise the end user experience to sate their own power trips.
51:35And so at the end of the day, I think either one of two things happens. Let's say you run r slash NBA and you're the mod and you choose to black it out. One of two things happens. Either some other enterprising person who wants to contribute their labor in exchange for power will start r slash real NBA or r slash NBA, like true NBA, and create a competitive product, again, in this marketplace of ideas and market efficiency, and ultimately end up replacing the same intent that r slash NBA captured originally. Either that will happen or Reddit, which is totally justified in doing so, will remove that moderator and allow other people to mob that same community.
52:18Again, what is in the end user's best interest? I think a lot of what's been written misunderstands the atomic value swap between moderators and Reddit and conflates moderator opinion with Redditor opinion. Moderator opinions are not Redditor opinions because their value swap is different. They're rewarded with power in exchange for their labor. They are not going to Reddit to be entertained. Redditors go to Reddit to be entertained, to engage in dialogue and become part of communities. But mods don't speak for the average Redditor. And I think what most stories that are written about Reddit miss is that mod atomic value swap is different.
53:09Let's say you were teaching a class about, okay, you've got product market fit. You've successfully built something that people want. Now you need to figure out what the best value capture mechanism is, that business model, product alignment, or whatever. And you're instructing class of people that had product market fit. What would you tell them to go do? What are the steps that they could take or the questions that they could ask or any activity that they could pursue to try to nail this? Because it seems so incredibly critical. What would you urge these people go do to try to solve that second challenge?
53:41This framework of user-generated content networks really only monetize via ads, and editorially generated content can only really monetize off of subscription. Define editorially generated, just like you pay someone to produce the content. Exactly. If somebody is being paid to create content, the only way that you can really monetize that is through consumer subscription. Otherwise, you're competing against UGC platforms, basically for the same user entertainment minute, and you're competing against these platforms that have a structurally better cogs. They pay nothing for the content that they're creating.
54:24UGC platforms can never monetize via subscription because going back to the medium example, their value proposition to content creators is distribution. And introducing a chilling effect onto that theoretical distribution disincentivizes existing and incremental content creators from contributing their content to the platform. But what they have is free content that they can then monetize via ads and effectively undercut editorial content. Conversely, editorial content, because people who are paid to create content are primarily incentivized by economic value, not potential distributed value and future potential, you can and should monetize via subscription.
55:15Twitch was the first board I ever joined. And when we invested in the company, they were burning a million dollars a month just on CDN bills. I mean, we had 14 data centers on the world and we had another CDN bill. Turns out serving video content is pretty expensive. You pair that with no one wanting to advertise against live content on the internet. Not misunderstood, just nobody understood it yet. Brand advertiser adjacency apprehension through the roof. We're staring down the barrel of this increasing cost center and trying to figure out how to make money. And there was one board meeting where the team proposed two monetization mechanisms.
56:02One was called Twitch Turbo, which was a monthly subscription. You would default to a HD stream 720p. This is back when actually the difference between 480 and 720p was meaningful. No ads, basically focusing on a better content consumption experience. And the other product that we would launch or were experimenting with was what was called subs, subscriptions. And interestingly enough, subs didn't offer a better content consumption experience, but it did allow you cosmetic changes. So it made your name a different color in the chat room of the channel you're subscribed to. You got access to a specific set of emojis that you could use in that chat.
56:45And you were supporting the streamer who you were subscribing to. The next board meeting after was so wildly telling. Twitch Turbo performed as you would have expected. One of the big issues is the Venn diagram of viewers on Twitch and people that used ad block was basically a perfect circle. So there was no utility in removing ads. But Twitch subs exploded, absolutely exploded. And it was clear that by the end of the year, the revenue off of subs was going to cover the entire op-ex of the business. That's the thing. They had to go through this intrinsic experimentation process of, okay, well, where is the platform delivering value?
57:33Where is their leverage to potentially extract value? Where's the attachment to different constituents on the platform? And what are products that you can offer that align all incentives? So Twitch Labs is such an elegant product because it aligns incentives across the board. Streamers can make more money. The viewers feel like they are more deeply engaged as part of the community. They get to self-express. And they know that they're supporting their favorite content creator. So they have that additional emotional payoff. And I remember thinking to myself, wow, this is a new business model that the internet has not seen before.
58:16And it's crazy that it only exists for live stream gaming content creators. When I saw Patreon, I was like, this is Twitch subs for the rest of the internet. What do you make of YouTube? I'm really interested in everything you said around distribution versus monetization. I know you've also written about the specialized strategy will always beat the generalized strategy. And so specialization is offer one or the other, don't offer everything. But YouTube seems to have succeeded to a great degree offering both. People blow up on YouTube. There's this inherent virality possible there, but people also make a living there.
58:53And I'm also going to ask about podcasts, which seem to be like the opposite of YouTube that they provide neither. Super weird. I wonder why you think that might be the case. But starting with YouTube, why has it been able to succeed in both? Or am I missing something? No, you're not missing something. And honestly, YouTube is a data point that really challenges my assumptions and frameworks as thinking. My best answer to the continued success and monopolistic position of YouTube is that by trying to offer both monetization and distribution, it opens itself up to being flanked on both sides. And I think there are small things that you can look at that examine whether YouTube is strictly better or strictly worse along both of those vectors to its constituents.
59:50For example, Justin Bieber blew up on YouTube, discovering YouTube, no longer on YouTube. You have this dynamic of your best users graduate from your platform. You aren't actually the best way for them to monetize anymore. I think there are plenty of examples like a Mr. Beast or other successful YouTubers where it works for them. They still make a lot of money on YouTube. But I'd be really curious if those YouTubers weren't making more money more lucratively outside of YouTube as a platform, away from the direct monetization from YouTube. So if you were to look at a successful YouTuber, how much do they, one, multi-home as a content creator?
1:00:34So create content on YouTube, but then create content on TikTok or elsewhere. And then how much do they encourage their audience to effectively meet them elsewhere? It's like Emma Chamberlain selling coffee or something. Exactly. Oftentimes, successful YouTubers will achieve enough distribution that it opens up more attractive monetization mechanisms for them to monetize their engaged audience. And if you think about ads, ads are super effective at monetizing mass, shallow engagement. Prank videos. Prank videos work on an ads-driven platform. But when it comes to deep engagement and narrow deep engagement, ads are terrible for that because you are artificially capping the monetization that a consumer of content can deliver to the content creator.
1:01:37This is why like nobody's incentivized to upload 30 minute deep dives into niche content on YouTube as the dominant form of monetization, or more likely, if you were to find a niche content creator, in every video description, it would be support me on Patreon. Because Patreon is a perfect way to monetize narrow, deep engagement. Or conversely, if you are a celebrity, if you have celebrity value, the best way to monetize celebrity is consumables. Arguably, the only durable way to monetize celebrity is through consumables. And this is why I'm a chamber and coffee or Mr. Beast Burger or Skims or...
1:02:17Jordan and Chews. Exactly. Fenty Beauty. And every time that celebrity has tried to monetize outside of consumables, failed miserably. No counterexamples. Shots. I think it was Justin Bieber tried to launch an Instagram competitor called Shots. Celebrity affinity is a one-time subsidy in user activation energy in an exchange and not a persistent one. Let's look at shots as an example. Maybe it gets people to sign up, but it can't perpetually subsidize an activation energy of consumption or content creation. Whereas if you're buying gin, you're buying a piece of George Clooney, or I guess that's tequila.
1:02:58Exactly. They're imbuing a consumable with their celebrity and brand. That's the best way I can articulate it. And so going back to YouTube, I think where I would really have to revisit my assumptions is if the most successful content creators on YouTube single-homed on YouTube and didn't try to monetize in other means. That would mean that it was the dominant strategy of both distribution and monetization, and they didn't want to diversify. But the thing is, that's not the way that they behave. And so it suggests to me that while it's an attractive value proposition for distribution and monetization, it is not strictly better than all alternatives.
1:03:43And there is a gravitational pull in both directions away once they achieve a certain level of success. Can you tell the story of the investment decision that you are most proud of having made? It's probably Twitch. Maybe it's because it was just the first time that I joined the board of a company and stuck my neck out to make an investment. But upon reflection, it was probably the first time that I touched ground on my own path to conviction. As an adolescent, I played my fair share of computer games and video games. I pulled my fair share of all-nighters playing Counter-Strike. And I think the insight that I had as an investor and user is that gaming content is entertainment.
1:04:30I think one of the questions that was asked at the time of the investment was, why do people watch other people play video games? Why don't you just play video games yourself? The answer I would give is when people watch Monday Night Football, why don't they just play football? It's because when you develop an attachment and an affinity to something and you can recognize what excellence looks like, or you can appreciate what excellence looks like, it becomes a form of entertainment in and of itself watching it. So I reached out to them. One of my friends went through YC. I got connected to Justin Kahn, and he introduced me to Emmett Shear, the CEO.
1:05:10I spoke to him and was like, hey guys, I really believe in the direction. I think it's amazing. I work at this small venture capital firm based in New York called Thrive Capital. We're a$40 million fund. At the time, I'd love to find a way to work together. Great. We're raising$15 million. I was like, oh, that's awesome and we cannot do that. And they went on to raise their Series B from Bessemer. And it was tough because we didn't have a capital product to meet their capital needs. Then we raised our third fund. It was$150 million. And they were my first call. I was like, okay, guys, we have the fund size.
1:05:49It would be amazing to be able to work together. So we engaged with them. They're running a fundraising process. Again, it's expensive to host and distribute this content. And this is before they had found the monetization mechanism. I go to my partners and we have the deal lined up. It was hard. It was really hard because we got stuck in the fundamental question of why do people do this? Is it valuable engagement? From my perspective, it was so obvious. People are engaged. They're deeply engaged. And engagement can be monetized, even though it may not be monetized currently. And I think the issue was this existential question around whether that quote unquote engagement was real or true.
1:06:35Actually, we didn't get there. I was in one end zone when my partners was in the other end zone and other people were in the middle of the field. And we just couldn't reach a decision. I had to call Emmett and Kevin Lynn, the COO, and I was like, hey, guys, I'm sorry, couldn't get there. I wish I had different news. I think we're out as a potential investor and wish you all the best. That night, I couldn't sleep. I stayed up until 4.30, re-articulating my thoughts, compiling why I thought it was a compelling investment opportunity, comping the minutes consumed against other platforms, Netflix, Facebook, all these other platforms where they were able to monetize that engagement and what it might look like at scale.
1:07:22I sent it to my partners and we talked about it the first thing in the morning. To everyone's credit, we were able to get there. We reversed our decision. I was so excited to call Emmett and Kevin back and be like, hey, actually, I'm sorry about yesterday, but we'd love to lead the investment. And we got there and we led the investment and rest is history. That's probably an investment decision I'm really proud of. It's the same story of the insight, the earned insight, same story that might be behind a great company. Totally. And I think I'm proud of myself because I didn't let go. I didn't let go.
1:07:53I felt like I put it all on the field. And honestly, conviction is such a rare bird in this industry. It can get snuffed out by process. It can get snuffed out by decision-making structure. And And for me, that was the first time that ever really pounded the table on anything. And it was the first time that I ever stuck my neck out professionally, but it felt right. How would you describe the state of the union of consumer investing then today? Because it seems like the case you made earlier is that absent a new major dislocation, maybe the vision represents that thing. We're deep into the saturation of things dependent on mobile, maybe even on cloud.
1:08:31The technologies that enable new stuff have been explored. It's mapped territory. And therefore, you probably shouldn't do digital consumer investing. Is that a fair summation of the consumer investing landscape today? Or is there something else that we have to say about it? Whether it's a hot take or a hard truth, I think that's exactly right. You look at the most recent, quote unquote, success in consumer. It was nine years ago. It was TikTok. It was musical. And nothing since. And this is what I'll call fast. the ability to create lots of enterprise value per year very quickly. As distinct from there will always be new consumer brands that are grown more organically, maybe they're bootstrapped.
1:09:14They shouldn't be financed with this kind of capital. There are plenty of examples. There's a business to be built selling widgets that doesn't require venture capital and doesn't require a wino. What does this make you do with your time? Because you're an investor curious about and interested in this space and deploying capital in this space, that's your job. And the opportunity set is left us a bit wanting for the reasons we just described. What do you do in the face of that? Do you change your interest? Do you pivot elsewhere? It seems like juicy existential problem. It does beg the question, what are you supposed to do?
1:09:48Barring an excellent answer to that question, definitely good use of time. I tend to follow of my curiosities. I like hanging out on the edges of things. I think the reason why I was so attracted to VTubing is not just that it enfranchises this new form of content creator, but it was also experimenting with a new modality of human computer input, motion tracking. It maps to, I think, the advance of the Vision Pro, which is taking motion tracking, maybe not to the extent of eye tracking, but definitely motion tracking as this fundamentally new human computer input. Based off of my framework of we need new human computer paradigms to create these new radioactive isotopes of capturable enterprise value.
1:10:45And the only way that we're going to be able to get that is if we have new methods of input into computers. I think that's what I have been looking for. I don't want to over-index on the potential of it, but it is really exciting to think of the possibilities that will emerge on the back of a new piece of adopted technology that increases the data write speed between humans and computers. It's almost incomprehensible to fully grasp what will be possible. We're limited by the scope of our imagination right now. What is your method for finding the most curious stuff? Is it just reps? Are you just constantly pulling on a thread and devoting real hours of the workday to just seeing what is going on?
1:11:32Or is it your personal entertainment? Are you watching CodeMiko? Or is it more that you're just, wow, this is a phenomenon that's interesting to me, but I'm not going to sit and watch it. I think it's a quirk of mine that I don't know what this says about me as a person. All of my quote unquote internet time or social media time, I spend none of it interacting with people I know. I spend no time on Facebook, no time on Instagram, no time on Snapchat. I don't even know if I have Snapchat installed on my phone. All of my time is spent on Reddit, TikTok, Twitter, other platforms. And I think what I like the most about spending time on those platforms is it gives you, let's call it a lossless view into humanity.
1:12:18It gives you these anthropological primary sources of what's going on. I just read Rick Rubin's... Creative actor. Creative Act, he describes it as this thing called source. And one could argue that is this concept of data out there to be absorbed. And I think for me, I spend so much time exploring on these platforms in search of source. I have my antenna trained, my radar dish is pointed into the sky, constantly looking for this source. And then I'll see signals that will suggest to me that there's something there. And then that will be the beginning of a thread that I start pulling on. Totally fascinating and fun.
1:13:04It's so fascinating. People are interesting. There's a company called Games Workshop. They own IP called Warhammer that has a tabletop miniature game. That's all they make. They make tabletop miniatures. It's a publicly traded company worth$8 billion. And they sent a cease and desist to a site that was hosting 3D models that ostensibly encroached on their IP. So for me, that signal in and of itself of, oh, there's this really large company trying to stifle this bottom-up motion, that blips. Pulls at your heartstrings. There's something there. It's things like that, almost countless other ones, where this person, they're a sophisticated actor.
1:13:49They are saying to themselves, this is economically worth my time to allocate in this way. And that is so interesting to me. And the more repeatable that is, the more that that is just a tip of an iceberg is the extent of my exploration of pulling on threads. Is all value creation at the end of the day just increasing convenience or reducing friction? And that's the story of technology just over and over and over. That's definitely a large bulk of it. I think that that doesn't meet some of the other needs around self-expression and self-actualization. If it was just homo economicus and not homo sapien, those would be the only two things.
1:14:32Better, cheaper, faster. We're not homo economicus. We are homo sapien. and somebody will pay more for a t-shirt because it says something about them than another t-shirt, even though from a utilitarian perspective, homo economicus wouldn't do that because they're fungibly the same thing. They serve the same utilitarian purpose. But again, going back to Leonardo Arnault is the wealthiest man in the world. Yeah, they increase friction to buy their stuff, I suppose. Yeah. Louis Vuitton handbag has functionally the same utilitarian value as any other bag out there, but it says something completely different from a social signaling expressive utility perspective.
1:15:07Yeah, maybe it's true for technology and not beyond technology. The handbags, the famous ones at Hermes basically look the same as they did 60 years ago. So it's much more about they make it easier for you to signal something about yourself. So it's still making something easier. You're willing to pay a lot of money for it. But technology seemed to always just be about friction reduction. That's what it seems like to me. And I'm curious if there's a dark side to that too. We talked a little bit earlier about the inequities that result from new technology. Naturally, technology creates more inequity, not less.
1:15:38How do you think about the dark side of tech and maybe with an example of what's going on in the world today? Yeah, it's tough. Technology tends to introduce power law curves of demand and adoption or exacerbate those curves where they previously weren't power law distributions. You end up with this dark side, the shadow that gets cast by the parallel distribution, examples of taxi medallion holders committing suicide because of the economic disparity created by Uber. They're so in the hole that it just doesn't make sense for them anymore because what they thought was a projected future cash flow was no longer based off of competition.
1:16:21Or we talked a little bit about mobile dating and Tinder and the effect that that has on the rise of incel culture. Anytime there is a new power law distribution that granted awards a ton of value to the head of that power law in the form of enterprise value capture. It just wipes out a ton of integral value in the middle of that curve and the long tail without exception. And so for every Uber, we have the taxi medallion holders, Airbnb, we have countless mid-market business hotel chains that have gone out of business. So the answer is yes, there always is a cost associated with that reallocation of value.
1:17:06I think there is a good amount of net new value creation anytime there's technological advancement, but there is also a ton of reassignment of value away from that mid to long tail to the head in that new distribution. I have one more question about the role that system settings play on investment opportunities. You've talked a lot about this why now idea and dislocations, enabling technologies, et cetera, and why they're so important. What about at a national level, specifically around the United States? It's amazing how disproportionate it's been, the percent of new innovation that's come from the US versus anywhere else in the world.
1:17:46Why do you think it is that basically all of the major innovations have come in the last, call it a hundred years or maybe even longer from the United States specifically? What is it about the system itself do you think has caused that to be true? And do you think it will persist in the future? This is a fancy way of asking, could you just invest in American-based companies? Gosh, the United States is such an interesting N of one. I don't want to come across as just hyper-nationalist here. It's truly remarkable. And I think it boils down to a few things. One is the United States, and this has been analyzed by people much smarter than me around the access to natural resources and the geography of the United States.
1:18:27But simplistically, from a system design perspective, the United States had a last mover advantage. As a country, It's so young, but it got to learn from all of the errors and unforced errors of the system design and governments from every other country. Turns out, yeah, monarchies, not good. Benevolent dictator structure, yeah, I guess that works only if they remain benevolent dictators and power corrupts, absolutely. So you get to design this incredible system of checks and balances. And I know we talk a lot about the founding fathers in the system design of the United States. And I cannot stress enough the elegance of the fundamental system design.
1:19:10If I were to distill down America's advantage at a global scale, it is that we have a monopoly on ambitious immigration, which is insane to think about. As a country, you get to skim the cream of the crop of hungry, smart, talented people from every other country. That is insane. And it exists because of this concept of the American dream, this idea of social economic mobility, the idea that you can leave ostensibly everything behind and come to a country that is a meritocracy and create a better life for yourself and future generations, that position is so dominant. It's almost this compounding moat of talent.
1:20:02There's a lot of people that say the United States is dead or it's end of the era. As long as there are lines around every visa immigration center, the US isn't going anywhere as a dominant power. I can't stress this enough to be able to just take every smart, hungry, talented person that wants to bet on themselves and absorb that ambition, that entrepreneurial nature, that enterprising nature into this country of immigrants, that's broken. That's such a broken advantage at a global scale. I forgot to ask earlier about a category, which is always fascinating to me and everyone, because they tend to be such good businesses, which is marketplaces.
1:20:44And this idea of how to start one effectively, which I've seen you write about. I can't remember exactly what you said, but something to the effect of if there's multi-skews in a marketplace, it has to start with an opinionated individual or something like that. I'm sure I'm messing that up. But I love this business model and therefore love any thoughts on how these things most effectively start and grow, in your opinion. I don't know if I have an answer to a repeatable motion to create marketplaces. Actually, let's define what a marketplace is, because I think there are varying interpretations of what a quote-unquote marketplace is.
1:21:15Loosely, I would describe a marketplace as where supply competes for the benefit of demand and supply does not engage in demand generation. Counter examples, Shopify. Shopify is not a marketplace. Shopify is not a marketplace because all of its supply does its own demand generation. Amazon is a marketplace. Amazon Marketplace because none of its supply does its own demand generation. It relies on Amazon for demand generation. I have a framework which is all behaviors that exist at scale in a lower friction format exist at subscale in a higher friction form first. So go look for that. Exactly. So I think a way to identify marketplace opportunities is looking at places where people are jumping through hoops, crawling through glass to meet liquidity.
1:22:24There's a lot that's been written about what makes a good marketplace. Bill Gurley has a great blog post of what makes a good marketplace and what are the different dimensions to measure marketplace. But at a minimum, I think that if there are primitive examples in the wild of supply and demand engaging in extenuating behavior to meet each other, that is a need to have before you're able to capture that in a lower friction scaled model. Marketplaces are so rare. Their rarity is proportional to how good of businesses they are. They're such good businesses that the efficient market is ruthless in attacking any and all opportunities that are exposed to them.
1:23:10And so barring, again, a dislocation in a market of a new technology or regulation, you could probably safely assume that if there was going to be a marketplace, Darwinistically, it would exist. You started the conversation a little bit talking about business models. Is there something that's the opposite of that? If marketplaces are so good that it's like the hypothetical$20 bill, someone would have picked it up already. Is there something where the business model is generally so bad that it's actually essentially an interesting place to apply entrepreneurial effort and creativity? Yes and no.
1:23:43One thing that I like to explore in the age-old debate between idea and founder, is there any founder that could make a company successful where they sell$20 bills for$10? I think the answer is no. No one in the world can make that work. It's just so fundamentally flawed from a business model perspective, that no amount of charisma, salesmanship, recruiting, fundraising can make that work. On a long enough timescale, that falls down like a house of cards if it gets stood up at all. Now, that's different from things that may be perceived to be irrational at first glance. So let's take Robinhood as an example.
1:24:20Robinhood gives away trades. Trading is free on the platform, which is insane if you look at the market pre-Robinhood. You had companies like Scott Trade, TD America Trade that were charging users$6,$7,$8 per trade. That was responsible for a third of their profit. And you had this set of incumbents who would never have thought to make their trades free. And a new entrant saying, we will make trades free is completely irrational at first order. And honestly, without payment for order flow as a business model, maybe Robinhood wouldn't be possible. Without the credit card companies being willing to pay for leads, Credit Karma wouldn't be possible.
1:25:08Credit Karma is like, okay, we're just going to give away credit reports for free. And Equifax and Transition are like, what are you doing? That's crazy. We're making so much money charging people for this. And Credit Karma is like, no, that's not what we're playing for. We're playing for something else. So it's possible to look at an ecosystem that's been relatively calcified and say, there is this opportunity to do what is considered first order irrational by the existing incumbents. If there is a greater opportunity to play for that is second order, there is a business model that is accessible on the other side, but only if you're willing to do this thing that is - Classic counter-positioning.
1:25:47Tegas did this really well with expert calls where fraction the cost, which seems stupid. Exactly. Exactly. But they make it up elsewhere through subscription. Yes. Could anyone have actually turned WeWork into a success? Or was it just so structurally plagued? Was it selling$20 bills for$10? How much of the industry misidentifies business models as value generative and sustainable when they're actually just selling $20 bills for$10. Can you talk about sculptors versus painters? It's an analogy that I reach for depending on the framing. I think about painting as additive. It's an additive process.
1:26:32You start with a blank canvas and your differentiation is the amount of paint that you add to it. Sculpting is a subtractive process. You start with a block of something and then you subtract from it until you find the thing. Michelangelo very famously has a quote, I saw the angel in the marble and set it free. That is describing what the process of sculpting is. In company building, in many things actually, you start as a painter and you end as a sculptor. And sometimes people miss that step. Founding a company is painting. You start with nothing. You have to bring 100 % of the things that differentiates you and add it to the canvas.
1:27:19But oftentimes what founders miss is the point at which they need to start sculpting because you can overshoot. Sculpting is a lot like management. You only become more efficient if you are able to cut things away and entrust them to other people. Otherwise, if you can't let go of things, you're never going to get anywhere deeper than the starting block of marble. There are some operators that need control over everything. They need to make the final decision and micromanage every single thing. And that is the equivalent of zero definition on a block of marble. What a truly defined, beautiful sculpture looks like is the courage to cut deeper than others.
1:28:09and to carve that out and be comfortable with it. And unless as an operator, as a founder, you can do that, you will just have this undefined blob that looks like nothing at all. It's a really interesting way to think about people too. There's probably inherently people that are more prone to one versus the other. I love these binary sorters. Oftentimes in life, we have to switch between the two. What does it mean for an investor? Are investors painters and sculptors? Maybe. I always tell people that to be a great investor, you need to have great taste in investing. The same way that we have taste in music and food and art.
1:28:51I think as an investor, you spend your formative years training, developing an understanding of what you like and what you dislike, what you're good at, what you're not good at, what you think you're good at that you're actually bad at, what you think you're bad at that you're actually good at. And then developing this felt sense and strong opinion. No three-star Michelin chef does anything that other people do. They have such a unique perspective. They have such a unique point of view and taste in what they want to put out in the world that that is what separates them apart. I think about the existential question of what companies, businesses out there am I the perfect investor for.
1:29:35I think it's an incredibly interesting place to wind what's been a fascinating discussion. I have one traditional closing question that I ask everybody. What is the kindest thing that anyone's ever done for you? It's gotta be my wife marrying me. Chris, thank you so much for your time. This is amazing. Thank you for having me. If you enjoyed this episode, check out joincolossus.com. There you'll find every episode of this podcast complete with transcripts, show notes, and resources to keep learning. You can also sign up for our newsletter, Colossus Weekly, where we condense episodes to the big ideas, quotations, and more, as well as share the best content we find on the internet every week.
1:30:34Thank you.
From the publisher
My guest today is Chris Paik, who is a General Partner at early-stage venture firm, Pace Capital. Before Pace, Chris spent eight years building and investing alongside last week’s guest, Josh Kushner, at his firm Thrive Capital. During that time, he sat on the board of Twitch. Our conversation explores Chris’ frameworks for investing, from atomic value swaps to business model product fit and the seven deadly sins. We also discuss different consumer trends like the rise of virtual YouTubers, Apple’s Vision Pro, and why everyone graduates off YouTube. Please enjoy my conversation with Chris Paik.
Founders Podcast
Founders Episode 136 - Estee Lauder
Founders Episode 288 - Ralph Lauren
For the full show notes, transcript, and links to mentioned content, check out the episode page here
-----
This episode is brought to you by Tegus, the modern research platform for leading investors. Stretch your research budget with flexible expert calls you can trust. At a fraction of the cost of traditional expert networks, Tegus customers pay only what an expert charges – with zero markups and no confusing call credits – netting an average 70% savings. Don’t want to conduct a full hour call? Tegus offers the ability to schedule 30-minutes, an offer you won’t find anywhere else. And they don’t stop there. With white-glove custom sourcing for every project and robust compliance measures, including a dedicated 50+ analyst team that vets every call transcript, Tegus ensures your privacy and protection. As the industry innovator for qualitative insights, Tegus helps you find the right experts you need at a quality and speed that can’t be matched. For a limited time, as a listener, you can trial Tegus for free by visiting tegus.co/patrick.
-----
Invest Like the Best is a property of Colossus, LLC. For more episodes of Invest Like the Best, visit joincolossus.com/episodes.
Past guests include Tobi Lutke, Kevin Systrom, Mike Krieger, John Collison, Kat Cole, Marc Andreessen, Matthew Ball, Bill Gurley, Anu Hariharan, Ben Thompson, and many more.
Stay up to date on all our podcasts by signing up to Colossus Weekly, our quick dive every Sunday highlighting the top business and investing concepts from our podcasts and the best of what we read that week. Sign up here.
Follow us on Twitter: @patrick_oshag | @JoinColossus
Show Notes
(00:03:00) - (First question) - The history of humanity as a data storage and transfer problem and how that relates to new kinds of computer interfaces
(00:14:47) - Investing strategies; knowing when to invest in new technology
(00:16:39) - Assessing the gray areas of a potential investment
(00:22:13) - CodeMiko and VTubing
(00:32:51) - An explanation of atomic value swap
(00:39:55) - The 7 deadly sins and the degrees to which each sin is monetizable
(00:45:05) - Evaluating the people behind new technologies
(00:50:03) - The Reddit Revolt - the current changes happening at Reddit
(00:53:09) - Business model alignment; the steps to take when determining a business’ value capture mechanism
(01:03:52) - The investment decision that Chris is most proud of
(01:08:14) - The consumer investing landscape today
(01:11:19) - Chris’ methods for finding new content, technology and potential investments on the internet
(01:14:06) - The fundamentals of value creation
(01:15:06) - The dark side of tech, how can it be more than just friction reduction
(01:17:22) - Chris’ opinion on why so many innovations come from the US
(01:20:37) - Defining market places and how to start and grow a successful one
(01:23:23) - Chris’ thoughts on opportunities with bad business models
(01:26:19) - A breakdown of Chris’ analogy “Sculptors versus painters”
(01:29:34) - The kindest thing that anyone has ever done for Chris




