E78: Why Benchmark is Different Than Other Firms with Sarah Tavel & Eric Vishria

28:14You know, COVID's behind us, but then you have the interest rates changing, what happens there. You have Silicon Valley Bank and what happens there in this like crazy weekend where everybody was wondering how they're going to make payroll. And these things happen now that no matter what size company you are, you're going to have to deal with these huge systemic shocks. And I think that makes founders care and really value so much more, not the firm, but who that person is that you're going to be partnering with. And so that's number one where people care now more than ever, not more than ever, but certainly more than the last 10 years of who that partner is that you're choosing.

28:56And I hear that a lot with the founders that I meet every day. And the second thing is that there's a little bit of like, oh, we're not in that craziness of having to make these very fast decisions. And then you have the founders who did make those decisions and warn other founders not to do it, to really get to know the partner that you're having on your board. And I think those two things coming together, and of course, so many of the interesting trends that are happening in the world of technology, has meant that we just feel like this is the type of time that we love to make new commitments.

29:36It feels like the last few years, there was a lack of discipline between the relationship between investor and founders. Maybe investors were a bit too submissive. Founders were raising all these crazy rounds, et cetera. And now is a return to back to basics. I'm curious how you guys think about today, sort of the proper pushback between the investor founder dynamic. I think that in any relationship that we're trying to create something or build something, you need some tension, right? Like if you think about like engineers working with product managers and designers, like you need push and pull, like you need constraints, like you need it to go back and forth.

30:16And through that tension, you get better results, right? And in times it's swung, like it swings too far one way or too far the other way, right? Like I think one of the challenges, for example, would be with the East Coast venture capitalists or the European venture capitalists, The financing for tech companies was so scarce in those geographies for so long that they had too much power. The investors had way too much power, and I think it led to suboptimal results. You didn't have the ambition or the dreaming that you would get. Otherwise, you're kind of getting forced to cash flow break even as a three-year-old company.

30:54That's not necessarily useful. And by the same token, I think it can swing too far the other way, right? And so I think it needs to be productive. But ultimately, like the relationship and the partnership with the founder is really based on like working closely together and trust. Because then I think you can all realize like, hey, we're pushing back or asking you a question or pulling on something with the same objective of creating the biggest and best company that we can. Yeah. And I think that North Star alignment is really important. To me, that's exactly what it comes down to is I think there's almost a positive selection bias.

31:32And I feel this all the time when meeting with a founder, there's a group of founders that they just want to build the best damn company they can build. The one that has the most impact. They have the ambition to build something as great as they possibly can. and those founders don't want someone to just come to the board meeting and clap. They want to know what does great look like when they're hiring that next leader. We always think about pulling forward the future into the present. So you're not just optimizing for that next round, but you're really optimizing for what is the end state of this company and how do we build a company that has the best fundamentals, escapes competition, has the culture that is going to have all the right checks and balances to really build something that endures.

32:28And so I find that I get most energized when we're pushing a little bit more. I think like, and this is one of the benefits of only making one or two investments a year is that we're more enmeshed. We're almost always involved in recruiting all the leaders. We know the founders really well, and we can start to see, okay, it looks like, remember that thing we heard in the back channel when we were hiring this person? I think it's starting to show up, you know, and like pulling forward some of those decisions. And when you make a decision 10 % earlier and you make the hire 10 % better, 10 % faster, or whatever it ends up being, those things compound.

33:12But oftentimes, it means that we're advocating for something that we see as important, but not necessarily urgent before the founder themselves might otherwise have seen it because there's so much else going on. And oftentimes, the last thing you want to do when you're a founder is think about adding something new to your plate, even though you know it's probably the right thing to do and it'll be the right investment and it'll help you two quarters from now, even if it's going to hurt you the next two months. But that work means pushing a little bit and advocating a little bit. And we hope it ends up bending the curve a little bit more for these companies and leading to founders that want that.

33:56One of your beliefs is that platform teams exist to scale the GP rather than get leverage to the founder. And I found that curious because don't you want to add as much support as you can or get as much leverage? You have assistants, you have principal and Blake. And my take on that is that you believe the costs of having the platform team in terms of all the coordination or sort of how it excuses maybe the GP to spend less time outweighs the benefit of getting the extra support for the founder. We are not a group of individuals that rest on our laurels. We're always thinking of refounding Benchmark, re-asking.

34:31you meet somebody who's a super talented talent leader, you know, should we? And I think what we realize is even if we had it, we wouldn't use it because we actually love the work of, you know, helping a founder, you know, find that great sales leader. And we think that we, by being involved and, and, and ourselves as, as partners, getting the reps over and over and over again, that we ourselves get better at advising the founder on the different types of engineering leaders or sales leaders or whatever it is, the network, the back-channeling, whatever it is. And then, of course, helping close.

35:13We want to be involved in that and fully integrated. And having a blind spot on any leader, I feel it. When I'm not involved with helping recruit one of the leaders, It feels to me almost like a little blind spot on the team, but I just don't know the executive as well as I know the other ones. And having that full, like really having the full experience with the founders, going through those recruiting calls together, I think ends up meaning a better relationship with the founder and our ability to be a better partner to the founder. And so if we had it, I think it would be the least used talent team.

35:54And so, you know, it's just not us. I think so much of the value of a partnership with an entrepreneur happens in these very few moments a year. I think there's two or three. It's not every month where the entrepreneur is faced with a key decision, a crossroads of some kind and what to do, what not to do, and how to change something. And by definition, it's not an obvious answer. Like, of course, they've listened to the podcast and they've read the blog posts and they've looked at all the VC content marketing and they've read the books of the companies. But like, but it's situation specific, right?

36:34It's market specific. It's company specific. It's personnel specific. And in those moments, the entrepreneur wants a sounding board, a partner to just navigate it, to try to make it, as Sarah said, like a little bit better decision, right? And the question is like, what do you need to have in order to be a really effective sounding board in those moments? Because I think all the value comes from those moments. And I think in order to be a really valuable sounding board in those moments, you just have to have the context. And you get the context by doing the work on an ongoing basis. And then in those moments, you can be really valuable as a sounding board, ask the right questions, help the entrepreneur have conviction or more conviction or confidence in the decision that they're making and their recommendation and asserts that bend the curve.

37:28And so I think that's so much of how we tackle the job. That's well said. I'm only 33, but even I've been amazed at how every few years there's just a new crop of entrepreneurs. It's amazing. Yeah. And kind of the board changes a bit. And I imagine that there's some percentage of deals for whom you guys have a unique perspective on that there aren't other firms competing for. And then there's some percentage of deals that you mentioned earlier, the six term sheets. They're very competitive. It's a very competitive venture landscape. And you mentioned that deal activity this year is up, but past few years has been down.

38:02And other firms do more on content marketing and they do more deals. And so if these next crop of entrepreneurs are, hey, their deals are funded by X firm and that firm also just is more out there, how do you guys think about remaining the most relevant or the most top of mind or the most favorite firm relative to your best peers? It's a perpetual challenge. There's no doubt about it. Like it's a perpetual challenge. There are six of us covering technology. Like that's tough. And it's a tall order. So I think that part's there. I think there are kind of maybe two comforting things in it. One is the entrepreneurs we do work with, we have to deliver for them.

38:46And because there are not that many of them, we really have to deliver for them. And we want to do that. And we're in sense to do that, I guess, is the point. And so I think in every competitive situation I've ever been in, it's the entrepreneurs that we work with that have been the biggest advocates that have tipped it in our favor every single time. And so I think we have to continue to foster that. The other part is you don't have to do all of the investments. Like you don't have to get every company. You don't have to do that. we aspire to be in the few that matter and that really turn into big, really special companies.

39:25But that isn't all of them. It's not most of them. It's not even a large percentage of them. So you don't worry about seeing every deal. Some firms need to see everything and they can't miss anything. But - Just can't. Yeah, you're not structured that way. Yeah, we just can't. It's just not possible. What we have to do is really double down on selection. Yeah. Selection and, um, and giving the founders a great, great experience. Um, in the last decade, we had a lot of part of it was the sort of economic environment, but also we had this, this philosophy of blitzscaling, right. Um, and particularly around sort of tech enabled businesses that weren't necessarily software first businesses.

40:03And there's dozens of them that raised tons of money. Some of them did well, some of them did less well. We work as one in your portfolio. I'm curious what you think the broader lessons are from these kind of tech-enabled businesses? And if you think venture will keep investing in them, or yeah, what do you think are the lessons that you've gone pretty deep in marketplaces, some of which are more software than others? I think it was Bill who talked about how the more optimistic, or the looser the funding environment, the lower the gross margins are in the businesses that get funded. And certainly that is a dynamic that you see a lot or have seen a lot and is very different now.

40:43I think the other thing that I think about a lot is, we had a CEO of a big public company come in. He used this expression that I love, which is that bad breath is worse than no breath. And he was talking about account executives, basically, that so many companies during the last five plus years, 10 years, the last part of this bull run, have been so oriented towards growth. And the bar has been set by all these companies where everybody's aspiring to the 3x growth. These numbers that end up meaning a few different things. They mean it's an orientation towards consuming more capital more quickly.

41:29It is a land grab orientation. And there's reasons to do a land grab. If you have a network effect, then you want to run fast. You know, you want to, the stronger your network effect, the further ahead you are than the pack, the more dominant you become. And then it's just a self-fulfilling prophecy. And so there are certainly types of businesses for which, like, that game is never going to change. And we love to play that game when we can do it. When the company, it's the right company for that type of orientation. But more often than not, you see this dynamic where people chase growth before the fundamentals are there, before they're able to support it with the right kind of organization, where it might actually, you know, in kind of a classic marketplace, the idea of going after 15 cities very quickly as opposed to really going deep and dominating one city.

42:24That orientation, I think, has really changed just because the funding environment, as interest rates go up, there's a very clear counterweight that happens on the company's ability to fund these early days. And I think it's going to be healthier for these companies long term in terms of, again, the bones and the DNA that they end up crafting. Sequoia famously, in, I believe, 1999 funded, was it a web van? That's part of it, too. It was a joint investment. Okay. I assume Averson was the only thing to say that the company didn't work out according to Hopes, but they thought, hey, it's still the right thesis, and maybe it's not the wrong time or the wrong company, but a decade and a half later, Instacart, and then that one worked out.

43:07And I'm curious, going back to tech-enabled businesses, there's a lot of real estate companies that have raised as tech companies, not just real estate, but other kinds of companies. And I'm curious if you could imagine a similar story in the future of like, hey, was WeWork the right business model for venture scale? Just not the right time or execution or whatever. I wonder if you guys are thinking about more discipline there. Or actually, no, these are good businesses. Software is eating the rest of the world. And they make sense for us. Just business models. I can totally say never, say never.

43:39Yeah. We looked at a bull semen company this last week or the last couple of weeks. Bull semen. Bull semen, like literally bull semen, like semen from bulls. Any personal time? Right. Well, this is the right, it's the right question, which is like, it's a, you know, I think this is like the amazing thing. Like we never invested in a transportation company until we invested in Uber. Right. Like we certainly didn't invest in a real estate company until we invested in WeWork. And, you know, we've made very few hardware investments over the last 15 years, but we have an AI chip company called Cerebers.

44:14And like, the point is that all these are capital intensive businesses, right? And so I'd say never say never. You're trying to like look for where the ground is shifting in the world, where there's a seam between like, between incumbents or something's changing that is gonna allow, create a crack of opportunity for a ingenious, scrappy founder to break through. That's what you're looking for, right? And sometimes it's just a granite wall and there's no crack and that's a bad time, right? And maybe you can use a lot of money and get a lot of dynamite, but it depends how thick the granite is. But other times there's cracks and if you look in the right way and you twist your head and you turn, you can kind of see a path through.

45:03And, you know, that's kind of what you're looking for. And those are the best actually in some ways, because it requires this like, you know, cocked head looking sideways and the sun just right to see the path, which means a whole bunch of people won't see it. And then sometimes there's just like a giant wide open path. And what happens in those cases is like, you know, 20 people will see it and like everyone and we'll try to funnel through. And like a lot of times that just doesn't work for anybody. And so like, so you kind of, it's a risk capital business and you want to keep taking those shots.

45:38And I think, you know, I think Instacart like did work and they seem to have done a really nice job. Yeah, kudos to them. Yeah, kudos to them. Just incredible job executing in what is a tough market. And then the last thing I was going to say is in markets like, it's particularly difficult when you go through like a ZERP environment, right? It's your interest rates. For a long time, capital is raining down. There's new sources of capital. That's one thing we haven't talked about, which is there's all new sources of capital for venture, right? That outside of the kind of traditional endowments and foundations, you have a lot of sovereign wealth money and other things that have come into venture, family office, which will, you know, some will pull back, but whatever, there was a lot new capital.

46:20And then all of a sudden, like, it changes and interest rates start going up very quickly. And of course, you have a global pandemic, which completely changes office policies and whether people are in the office or not in the office. And so, like, I would also say, like, those are all like, they're not small complexities. Those are like societal changing things that have happened that make it particularly difficult in those kinds of businesses versus a traditional software business where your high gross margin, your primary cost is headcount. Those are much easier to adjust or turn the dials on, and especially in the recurring revenue business.

46:58And so some businesses are very forgiving to those types of macro changes and others are not. So what's the Bullsemen company do? Super interesting company, amazing founders solving a real problem. And it's kind of one of these ones where it's obviously way off the beaten path for us and not something we know a lot about, but you have to look. What do you guys think about when you think about the companies in your portfolio? Like what are companies that scream like this is a benchmark company? This is the type of investing that we want to be doing. You know, I think about Chainalysis, which was my first investment at Benchmark.

47:33What was so fun about that company is like, you know, number one, it's kind of the classic hearing about a company, having a little bit of a thesis on the area, just kind of being intellectually really interested, provoked by what was happening in crypto, trying to figure out and think through. You know, we talk a lot about this idea of, you know, seeing the present clearly and looking at all the use cases and not really having a lot of clarity at that point on what the use cases were going to be in crypto beyond, you know, what was happening at the time, which was store of value, raising money, more illegal things or outside of the normal legal framework, reaching out.

48:18And while everybody else at that point in time was thinking about ICOs and token investing and all these things, here we saw a company and a set of founders who saw something that other people didn't see. Michael, the CEO, had been at Kraken, one of the exchanges, and realized that if crypto was going to become a big thing. There was going to need to be a company that was going to partner both with the public sector and the private sector to get the public sector more comfortable with the private sector that was regulated, was touching fiat to then participate in crypto as well. And so he saw something before other people did, brought in two incredible founders.

49:03And then And we saw both the early traction momentum and a possibility for this company to become the dominant player in their space, which is rare to find in software as a service companies. And it wasn't actually at the time that competitive of a deal. The company was over here, this exception to where all the momentum was in crypto at the time. and we got to partner, be the first board member, which is always our aspiration to partner as early in the company's life as possible, ideally that first board member. And now it's been almost six years of partnering and hopefully many more to come.

49:50My first investment at Benchmark was Confluent and it was the three founders and they were like coming out of LinkedIn. They had started this open source project called Kafka. And that was it. And they were still employees. And so it was the initial round of financing and backing them. And it was a popular open source project, but people viewed it as kind of this like component that wasn't that important. That was a pipeline to something else. And the founders led by J. Kreps painted this vision of why like that view was incomplete. And it was actually going to be a much, much bigger thing. And here were the path that they were going to take to it.

50:34And it was like, you know, going back to this like granite wall analogy, like, you know, cock your head and look right through and here it is. Like, this is how it's going to go. And we have a lot of history with a lot of open source companies going back to Red Hat. It was like data infrastructure stuff that I like and really authentic founders who just like lived and breathed the problem, the solution, had thought about it extensively. And so, what a cool project to get to be an investor and participant in. And here we are now, I guess, nine years and it's a whatever, $10 billion public company.

51:12So from nothing to that over the course of that time has been a spectacular journey. And I think crawling through that crevice was something that they they did astonishingly well. Let's go deep on some sectors. I want to start with the consumer social. So you guys have done some of the more iconic consumer social companies of the past, you know, 20 years. I mentioned Snap, Instagram, others. And I remember we were talking a few years ago, before the pandemic, that we're saying, hey, maybe there's an opportunity to disrupt some of these giants. You know, people are bearish, but just maybe there's an opportunity.

51:45I'm curious for your reflections on what's happened. Let's say I went in a coma, you know, right before the pandemic or when we were chatting. What are your learnings or what happened in consumer social? This would have been a good move to go into a coma. I mean, look, I think you and I talked about a post I wrote about participatory social and this idea that I still believe in, which is that so much of what social is today is this lean back experience where you kind of just feel bad about yourself. It's one of the things I always loved about Pinterest, which was that it felt like this space you could go to online, which was rare and wouldn't make you feel bad about yourself.

52:28It was about dreaming and being inspired and thinking about a more creative life. But so much of social had become this space and we see it in anxiety levels and depression levels of young people. It doesn't make people feel good about themselves, but what makes people feel good about themselves is participating, engaging with their friends, doing that work. And so, you know, I still believe that there's going to be a company that comes along that maybe it starts off looking like a game, but will be about that connection that will emerge and will have the potential to disrupt these incumbents.

53:05But what happened instead in like, you know, the last four or so years was you had these two things come together, which was this real maximization of the algorithmic feed. So you're bringing in this kind of global maxima of content personalized for each individual, which makes these experiences so much more of what they already were. And then you also have the most addictive format that we've had in social, which is short video. And so, of course, those two things come together with TikTok, which is just consumed minutes. And Instagram is doing a really good job doing something that I've never seen another social company do, which is take the same atomic unit of another platform and establish it in their own and make it happen.

53:56But so they did it with stories and they're doing it now with short video and reels. And it's just so addictive. I read this book, Dopamine Nation, went on a social fast. And the thing that I just found that most pulled you in was short form video. And it's everywhere. You can't get away from it. And it just means that the job of building a social network was always really hard, right? Like the list of companies that tried to do it and haven't succeeded is exceedingly long. And then that level of hardness just got turned all the way up. And so it's tough. But I think it's kind of back to looking for the exceptions and a little bit not giving up when everybody else has.

54:42Continue to be open-minded here, but it has not been easy. There's always this hope that every new generation has its own social network. And so hopefully it's not just the same as forever. But it seems like LinkedIn is just a professional network for every... It feels like you mentioned the graveyard of companies have tried to compete with it. It feels like there's just no disrupting that. Is that something you believe or is that something you've looked for? I mean, certainly I've seen a lot of companies that are... We're the mobile first, Gen Z first, LinkedIn. I think the companies that I've been more interested in are actually the marketplaces that have labor and are building a profile for those workers.

55:27So we have two companies in our portfolio that I can speak to. One is InstaWork. I mean, this is just an incredible founder who has had an extraordinary level of grit because it was a marketplace for workers, for events. You know, imagine you're hosting an event and you need a bartender, the waiters, waitresses. They are a marketplace for that. It was pre-COVID. You can imagine this company was growing like crazy. COVID happened. Nothing. Really took the company down to the studs, launched in a new vertical industrial warehouse use cases. And now, of course, it's doing incredibly well having these two verticals going back to events and then the industrial use cases.

56:10And every worker, you know, if you're a bartender, are you on LinkedIn? No. But you have a profile with InstaWork. And the history of your work is on InstaWork. The other example is a company called Medli. It's a marketplace for nurses. You know, you're an outpatient surgical center. And if somebody calls in sick, you need a nurse, you know, pronto. You use Medli for that. And then you start realizing, this is what happens with InstaWork too, that instead of having the people there on staff, you can actually rely entirely on this new network of workers to fulfill all the work that you have. And same thing, these nurses have profiles now on Medley that is a reflection of their history as a nurse.

57:01And so these kind of vertical networks that have emerged, I think there's going to continue to be other verticals. Real estate always seems like one of the obvious ones where you're going to see these types of networks form that way. And they never look like the last generation, right? This is not a network. It didn't start with a network and then become recruiting. It kind of started with the human capital and then became a network. Who knows what the next generation will look like? The question I've been curious about for many years now, but I've never figured out the right solution for is how you bring reputation online.

57:35We've done it with certain categories like rate my professor, things like that. But reputation is so valuable. I mean, reference checks that we do is just so redundant on so many levels. And at the same time, it also feels like information that just doesn't want to be on the internet or something. But I think at some point, someone's going to figure out an elegant way to represent that because it seems like one of the great scarcities in this new AI era is information that's in people's heads that's not on the internet. Yes. Rich Barden was the CEO of many incredible companies, just in Expedia.

58:06Zillow was a venture partner with Benchmark. And one of the things he always talks about is the data that is in the shadows, whether it's a data system that's already kind of exists, that's structured, and you can bring it online or in our brains, and you can bring it into public domain, Glassdoor being an example of that, another company that he helped found. And that is always a recipe for really a big opportunity. Yeah. It feels like, you know, we've been talking about different eras, internet era, mobile era, you know, we had a bit of a crypto, you know, we're an AI, we'll get to that. It feels like different eras, you know, values accrue differently between incumbents and startups.

58:44It feels like internet era produced, you know, literal trillion dollar companies, mobile era, you know, hasn't been as long, so maybe they'll get that big, but it feels like they won't. It feels like they, you know, produce some epic companies. We've talked about some of them, but it feels like it was an extension of the internet. It actually enabled the internet companies to get to a trillion dollars. And it feels like that's sort of the consensus on AI. Then in some ways, it feels like an extension of something that will continue to compound, that there will be big companies, but the biggest gains will be reaped by the incumbents.

59:13Is that an accurate read? Or do you share that read? Or how would you edit that characterization? I think we don't know. I think we don't know. I think it's early. Like, I think in the AI world, it feels like there's kind of three distinct things that are happening, and I'm sure many others. But one is the model piece, right? And the foundational models and everything else. And I think that that's an area where there's a few organizations who are spending gobs of money. And nine months later, there are open source alternatives that are almost as good and maybe even better in specific applications.

59:48and like were developed by, you know, three guys, a dog and a couple million bucks. And so like, it's kind of an astonishing element and it's such a tremendously depreciating asset like that, the foundational models. There's a lot of cool work happening there. You know, we've always been long open source. I think that continues to be really interesting and there'll be a lot of cool stuff happening there. Second part is something that Sarah has talked about before, which is like there's new vertical companies where they're selling the work. A lot of it's selling entry-level white-collar jobs, whether it's medical transcription or paralegal work or assistants or analysts and so forth.

1:00:32LLMs in particular are really good at that. There are new companies that are developing to do that. I think there's a real question as to the durability or defensibility of those companies, but they're super interesting. They're getting real traction, They're solving real problems. So I think that's the second category. And then third is what you mentioned, which is their incumbents, whether it's GitHub with Autopilot or Copilot or Microsoft introducing the writing capabilities into the Office platform and so forth, where there is enhancement to an incumbent base product and distribution, which seems to work really, really well.

1:01:08And I think in that third category, AI has the ability to actually reshuffle the deck. There's some hierarchy of those products and those companies. And you could totally imagine the fourth or fifth place player or the startup who kind of comes in with better AI or more AI-centric approach, leapfrogging the winner just because it becomes so necessary. And I think you're going to see a lot happening there. You know, I think we're like watching all three and we're in it. And where does value accrue? Like, are we going to see some breakout companies in that second category in those vertical applications?

1:01:45It feels like we will. It's hard to decipher right now, but like, I think we're meeting them all and like trying to understand that and develop a conviction one way or another. And then the same kind of thing with the incumbents. And with the models, maybe I'm a little more skeptical right now. Just it feels like amazing technology. And I'm excited to see open source continue to win. But there's a lot of work to do to turn those into monetizable opportunities. Oftentimes, the first wave of companies that emerge can create a head fake in a way for people and how they analyze the industry. And so it is natural for the first wave of companies to kind of continue the mental model that we've had for the prior generation.

1:02:37And I love this saying, I forget who said it, which is like, the startup has to race to get to distribution before the incumbent gets to innovation. And certainly, like, a lot of what has happened is that you've seen startups that, you know, saw GPT-3, took advantage of it, and ran really quickly. But, man, the incumbents are really good at innovating right now. And so they caught up very, very quickly. and that created this kind of sentiment in the ecosystem of like, oh, the incumbents are going to be the main beneficiaries. But now we're starting to see this next generation. And I think a lot about what Eric was talking about with work as an example, where you always have to think about what the jujitsu move is going to be when you go against the incumbents.

1:03:27Like, you know, in marketplaces, actually, it doesn't really matter how big, you know, the incumbent is. It's always like, how much better are you at making the customer happy than the incumbent can? And what's a jujitsu move here? Like, certainly there's like new surface area, right? Like going after verticals that wouldn't have otherwise supported a software company that had to sell a productivity lift, you know, 10%, 15%, whatever, trying to make that business model work versus a company that's actually just selling work, which is a 95 % productivity improvement. That's very disruptive. It's new surface area that I really believe is going to create a new generation of companies.

1:04:10And then when you think about the incumbents, Eric's third category, well, where are they vulnerable? They're vulnerable in that most of them charge per seat. And that is a real innovator's dilemma. And so, yes, they can make each of their seats more productive, give them superpowers, but it might be that a company comes along and doesn't need to charge by seat, has a different way of thinking about what they're building, the value proposition, how they price it to the customer that is disruptive. Adam D 'Angelo, the CEO of Quora, another company that, another founder we get to work with, talked about this as the next industrial revolution.

1:04:54And it really feels that way. But there's something unfair with like, okay, if you're a trillion dollar company and you leverage AI in order to build 10 % more market cap, that's a lot of money. And so there's kind of a little bit of that dynamic that we all face. But when you're thinking about it from where we are, there's going to be some really important companies that get built. Yeah. It is interesting. You mentioned GitHub. GitHub was a company that had very low revenue when it sold. Is that accurate? I think it had a couple hundred million, 300 million. Oh, really? Yeah, not too shabby. But I just saw a hugging face to raise this big round.

1:05:28It feels like they have pretty low revenue. It feels like there's some companies that raise very high valuations, that have some real value there, really strong community, really strong data, but just haven't figured out monetization. And it feels like, I don't know if are built to sell to incumbents, but it feels like it's unclear how they become independent, you know, massive businesses. There's definitely a hype cycle in AI right now. And so I think that's just the reality. It always happens. It's fine. But here's the thing about hype cycles. Normally in one of these hype cycles, there are a handful of companies that like you almost could pay whatever and it still works as an investment.

1:06:04Like there are some like amazing things. And, And part of the reason there's hype and all these companies get marked up at these incredible prices is because everyone's looking for them. Everyone's looking for one of those companies that is going to be so special. And so it turns out if you overpay for a company that ends up being a trillion-dollar company or a$100 billion company, it doesn't really matter. That's fine. And there's always a lot to figure out. People forget the early days of Google. They didn't have monetization. They had amazing, amazing growth, right? Yeah. In terms of, they were stealing share.

1:06:38This is, I'm talking about like 99, 2000, I think 2001 even. They had just stealing tremendous share from other search engines because it was a better product. And they were really unclear on how to monetize it. I think at one point, they were actually worried that they were going to run out of money because they were spending all their money on data center costs, like to keep it up. And so, you can figure it out, like is my point. If you have enduring value. If you have enduring value and you have something that is there, it gives you the license or the opportunity to figure it out. It gives you the freedom to figure it out with some period of time.

1:07:11And, you know, that was always one of the big pushbacks on social networks, right? For a long time, if you remember, it was like, how are these social networks going to ever make money? Are they ever going to make money? Do they have a business model? Blah, blah, blah. And then like, here we have Facebook, which is what? 500 billion, 600 billion right now. And so like, you know, it turns out like it can work and can figure them out. What we've seen in this era so far is that companies can grow insanely fast or much quicker, can get attention quicker, but they can also lose that quicker as well.

1:07:39And there's this whole sort of meme at this point of just wrapper around GPT or people, I think Jasper is an example of a company that grew super quick and people are unsure about at the moment. How do you think about, I think you wrote something to this effect, Sarah, of what are the types of companies that are going to have enduring value or truly defensible? When you think about most SaaS companies, the reality is that very few of them have had a tech moat, right? Sometimes, occasionally, I know with Amplitude, as an example, they did work on the core database to deliver the service in a way that wouldn't have been cost prohibitive for them to do it.

1:08:16But very rarely are these companies actually really about some special IP. It's always about what is the advantage of scale that you get, the getting at the market faster, X with a better product, the work that you do as you build more and more of the features that people need so that you can deliver more and more value to the customer. And I think certainly you see that a lot, even with Jasper, of the work that they do, fine-tuning, getting more enterprise-ready. These things, you kind of take it for granted, but it's like real work that has to happen in these companies. But one of the things that I think about a lot also is, is there something that as you build it, you're building something else that has accruing value for the company?

1:09:08You know, whether that's the data set that you're building that lets you fine tune and make kind of the value proposition stronger and stronger, whether you're creating a new system of record for something, whether there's a new data asset that you're creating that's almost an exhaust from doing the core work, but is itself valuable. And there's also just like the last thing, which is the core enterprise software work that sometimes has to happen of like building the functionality that makes it easier for these companies to grow in an organization. And so, you know, the thing that I worry more about is trying to get distribution before the incumbent gets innovation.

1:09:49That just seems, you know, there's a lot of companies that were, you know, a new document type that you could use to, you know, use a large language model to help you write. and then Notion adds something exactly like that where you already are, it's really, really hard to compete with that. You're not trying to do that race against innovation for the incumbent. You're building something new that is a new surface area. That feels like the really important thing to focus on. I think the nature of moats, like moats are much more complex than they seem on the surface, I think. And I'll use it, I'll pick a controversial company, Tesla.

1:10:34And I think one of the things that's interesting, so a large car manufacturer, CEO said this, an incumbent car manufacturer, internal combustion engine one, said like, you know, it's like, what's the moat for one of those companies? And his view was that the moat was like how they build a car. And his primary concern wasn't like electric versus like this or like that. It was like, did Tesla figure out a better way to build a car? And what was meant by that was like everything, like the unibody design, the biggest casting machines in the world, the emphasis on like software battery development, like, and how do you manage that?

1:11:18And his point was like, every single thing that they had done over the last hundred years was built around one way of building a car, right? And an assembly line where you do things a certain way and you go this way. And Tesla may have figured out a better way to build a car. Like that's the moat. And that's the thing that has like shifted. We're trying to figure it out. Like where is the moat on a lot of these AI companies and however you think about it. But I think it might be more hidden and under layers or really like more convoluted or more complex, maybe than it seems on the surface. Say more about where you think people kind of misunderstand moats in general.

1:12:00Is it that they see them? Well, like you hear like these kind of simplistic things which are like data is the new oil and data is the moat. It's like, is it? Like, you know, like, okay, like maybe. Or like, you know, one that's popular around the AI companies right now is like access to compute or whatever. It's like, is it? Like, I don't know. And those feel like things which are more transient or their way – I think what – take data as an example. People have very consistently figured out ways around that. And I'm sure I'm wrong about some other case. Like there's some case where there is a huge proprietary data set that's really valuable like Rentech, right?

1:12:42Renaissance Technologies, they have built, I think, their business, they would say, as a fund, an investment house, is their data moat, which they have built over however many years and really took time to develop. But in general, I don't know that most of these companies who think they have a data moat have a data moat. I don't think there's that much value in their proprietary data, it turns out. Yeah, they have their data, and someone else has data that's available on the internet, and it's basically as good. So they're defensible for other reasons? Yeah, it seems like. I mean, it seems like it, right?

1:13:14And so I think you kind of keep going through these things and you see examples like that where you can kind of come at it from a different direction. Let's take another example just because we're talking about these really interesting businesses. It's like forever people thought like with Netflix, like their back catalog was the moat. Yeah. It changes. That's really interesting. Venture capital firms don't have data moats. No, we do not have data moats. We benefit from strong brand, like long feedback loops. So brand becomes a moat. Yeah, I think brand is a moat. And then you just have to wake up every morning running with absolute reckless abandon hustle to maintain that.

1:13:54Peter said that the day that I, Peter, don't want to cold email a founder or think I'm too good to do that is the day where I stop being a good VC or something like that. It takes that. It really is something to that. Okay, we talked a bit about AI. I want to talk about crypto as well. Here's my read from the outside on benchmark crypto. You guys made some great, you know, you made some money, made some great bets, chain analysis, you know, Bitcoin early, Zappo, I'm sure. So rare. Yeah. Yeah, a couple in the nation. I'm sure, you know, a few others, but relative to your peers, I don't want to say you sat out, but you didn't go all in in the same way.

1:14:25You were disciplined, relatively disciplined. And the firms that lost the most and maybe, you know, won the most feel like they were just more comfortable YOLOing. And it feels like that's not as much in the benchmark DNA because you're more fundamentally more sophisticated. You're more going to do the thing that guaranteed does well versus willing to take these kind of like, you know, bet it all type thing. Is that an unfair read? I'm sure part of it's YOLOing, but also part of it is it's just kind of like a different kind of fundamentals, which is more like what are other buyers going to do?

1:15:00or more about like, you know, even NFTs is such a status-driven thing, more like sort of awareness of the markets and just like a different kind of investing than fundamental venture investing. Yeah, I think that's what it comes down to. It's actually like the way I think of it is like, there was a class of companies playing in the crypto space where they were founders that believed that company building was going to be really important. They were founders that We're excited about the technology that, you know, is getting more and more mature around blockchains and crypto generally, but also recognize, you know, are oriented towards building an enduring company, doing the company building.

1:15:44I think we have a very large percentage of those companies. you know the founders of so rare are just incredible and from the very beginning even though they were participating in this nft world that had a lot of hype around it they were very oriented on finding a found a partner to help them build a really enduring company same thing with global illumination which we just sold to open ai and then you know chain analysis wensys with zappo yeah um and so those companies those founders were very oriented towards those types of businesses. Then there was another class of company, which is where the crypto focus funds ended up really orienting towards that were very focused on the example of Uniswap of a company that built an extraordinary amount of value with a super, super small team because they were able to use tokens and token economics in order to not need to do the company building and to create the right incentives to enable and facilitate that ecosystem.

1:16:50There were great examples of companies there, Uniswap being one of them. There are many more. And there were a lot of companies, kind of to the earlier conversation with Eric, that it felt like a very, very short-term orientation. I felt a little bit like it was the culture that kind of got set in crypto with ICOs, where you write the white paper, you raise a lot of money, and then you'll figure out how to build the thing. And the founders at that point, they don't even need to be around the hoop because they've already got their tokens from the ICO raise. That permeated so much of the culture of building in crypto.

1:17:32It permeated how people thought about NFTs. And that just isn't our orientation. And like, it's a game that a lot of people were able to play extraordinarily well and like, kudos to them. But like, that just isn't who we are. Like, if we wanted, there are ways that benchmark, we could all make more money, right? Like, you know, we were joking before we started about the benchmark growth fund, like, you know, wealth management, all those things. Like, there are always those types of opportunities. And certainly, that era felt like it for a while. I joked once that I felt like I was a professional basketball player and I took a rocket ship to Mars and I tried to shoot a hoop and then the ball just floated away.

1:18:14There was so much that wasn't who we are and the game that we want to play, which is that building enduring value, really partnering with the founder who wants to do that. And that's why we're here, is doing that work and that type of partnership. And that just felt very different. And so I think we, of the companies that fit the mold of partnership that we want, we were there. We were in that arena. But we just, the other thing, you know, people did a lot of great investing, made a lot of money. That's not, that just wasn't, that type of orientation wasn't our thing. We're all in the arena just trying stuff.

1:18:56Some will work out. Yeah, I can't believe I said arena. Yeah, exactly. Let's talk about healthcare for a minute. Uh-oh. Yeah, I was wondering if things have changed because we talked a few years ago, you know, your wife is a long time healthcare entrepreneur and investor and you were bearish about it after going a little bit deeper as a firm about it, just it's so hard, et cetera. Nothing changed? Not looking at something. You know, healthcare, health tech, it's just, it's a really humbling space for a founder. And like, you know, there's a number, I met with a founder just the other day, so super talented team that, you know, one of these, they had an idea using AI for, for a value proposition that makes so much sense.

1:19:44And then you start to ask the question of like, well, who's going to pay and how is the, where are the incentives going to come from for that to happen? And you're just like, oh, these poor guys, like they're going to, I introduced them actually to a friend of ours, Malay, because I was like, Malay, these are super smart guys, but they need help guiding the topology of healthcare and just incredible incentives that are so not what you would rationally design if you had a blank sheet of paper. And so I think it's one of those spaces. There's some spaces in venture and what we do where I joke, you have to marry them, you can't date them.

1:20:29I actually felt kind of crypto token world was one of those spaces where if you try to date it, it's going to be really difficult. It's so specialized. I think healthcare is one of those two where it's a really hard one to date because you just have to have that mental map in your mind of all the incentives and all the different players and how it all fits together. And it's really tricky to be surfacing there. To take the contrarian view. Please. There are emerging areas. You're on the board of Medley, Nurse Marketplace. I work with Benchling and a company called Acuity MD, which serves the medical device companies.

1:21:07And so there are these areas that are not directly hospitals and patient care. and the weird triumvirate of insurance patients and providers and everything that goes on between them and the weird incentives between them. But actually, all of these other areas around them, which really do contribute to a better patient experience, better medicines, better patient experience through better medical devices or better nurses, better drugs that are important, there is a lot of interesting innovation happening and we're actually bizarrely invested in them. That's true. And, and, and actually all those companies are doing well, which is kind of interesting too.

1:21:55And so I think there's this other dynamic, which is just like, maybe the way to tackle the industry isn't to mess with this like triumvirate, which is at the core of it or have to integrate with the EMR, which is, you know, controlled by one company largely that has like literally no incentive to do anything except increase the percentage of GDP that we spend on healthcare in the country. Like, it's just like, maybe like, maybe we just have to come at it like this away, you know? And I think we'll continue to see those and excited to see those companies because there is a lot of spend there. There's a lot of waste there.

1:22:37There's a lot of room for innovation. And so it's just like finding the angle of attack. I appreciate the light disagreement there. I said as a segue to when you guys are debating sort of the refounding of the firm, what are the debates that keep coming up or what are the conversations or questions that you guys are asking yourself when you think about, hey, I know we've said we're gonna do this for so long, but is this something we should consider? What are the questions? Probably the one that I think about the most in part because I grew up in the industry this way. I started my career in venture as an analyst at Bessemer.

1:23:14And so lowest rung of the ladder, cold calling founders, which was kind of an early idea. Jeremy Levine started that program at Bessemer and now has become the default in a way. Like these other firms, as we talked about earlier, they see 100 % of the companies that are getting founded or funded. And one of the conversations always, we have chosen thus far to be really the six general partners. We have Blake, who is incredible as a principal. I think we've had maybe five of five Blakes and benchmarks, 25 plus year history. And so the paranoia that I certainly have all the time is there is going to be that founder whom mistakes our lack of outreach for a lack of interest.

1:24:10We are constrained by the fact that we have six people. And other firms have dozens of people who are reaching out to seed funds, cold calling the founders. And so it's very easy for a founder then to kind of look at their inbox and decide whom they're going to raise from based on their inbox. I actually had a funny experience with a friend of mine who's a YC founder. And he was telling me, he's like, my process, you know, it's just not moving that well. And I was like, well, tell me who you're talking to. This was like pre-revenue company. And I looked at his list and it was all these like multi-stage venture firms who had reached out to him.

1:24:52I'm like, oh my God, like, it's like, it's a wrong signal, but a lot of people make that mistake. And so, you know, there is a corresponding, you know, the trade-off, of course, to having more people at Benchmark besides the GPs. One of the biggest ones is just, does it end up atomizing our attention away from the collaboration that we have with each other? And there are many, many other things to think about, but that is certainly when you ask me where I'm most paranoid, I want to make sure I'm seeing the best companies that I can possibly see getting to meet those founders. People tell stories about us.

1:25:33Totally. They move whatever. They're going to want a lower price or they're going to be slow or whatever. People tell these stories. They only say yes once or twice, whatever it is. We need to work against that all the time to meet those founders. and other firms have developed machinery around it that we don't have. And that's a trade-off. It's interesting because in a competitive market, you don't want to be, no one wants to be like a worse version of other firms playing their game. You want to play your game to the max. A hundred percent. And so, yeah, you could have a little bit of platform team, but these firms have masses.

1:26:11No one's ever going to compete on that level, but what they're not going to compete with you on is GP time and attention and focus. And so maximizing that. Yeah. And it's a great, like, you know, one of the trade-offs is like when, you know, we cold email, cold DM people all the time, right? And, or, you know, do the kind of the first call. That is us reaching out. There's no army of people who are doing it. It is us. And that, and like, there's no person when we engage with a company, there's no person to whom I say, oh, like reach out to the, you know, these other companies or do these diligence calls or go through the spreadsheet if there's a spreadsheet that rarely is.

1:26:52But that's us. And again, we think that leads to a better experience for the founder. It leads to better decisions by us. But, you know, there are trade-offs to everything. Yeah. Benchmark does the work. That's a great place to wrap. Eric and Sarah, thanks so much for coming on the podcast. Thanks for having us. Thank you for having us. Yeah. Turpentine VC is a podcast from Turpentine, the network behind Moment of Zen and Econ 102. If you liked the episode, please leave a review in the Apple store or rate us on Spotify.