Uncapped #22 | Greg Rosen from BoxGroup

Greg Rosen is a Partner at BoxGroup. Greg was the first hire at BoxGroup outside of the founders, David Tisch and Adam Rothenberg. After moving to the West Coast to work with Benchmark and Bedrock, Greg rejoined BoxGroup and currently invests out of their San Francisco office. An engineer by training, Greg built iOS games in high school before dropping out of college at 19 to join Jim Pallotta's venture fund in New York City. BoxGroup is an NYC-based seed stage venture capital firm that has invested in over 500 seed-stage startups over the last 15 years, including Plaid, Ro, Ramp, Clay, Scopely, Warp, Cursor, PillPack, Amplitude, Flatiron Health, Stripe, Warby Parker, Harry’s, Oscar, Flexport, Classpass, Vine, GroupMe, Airtable and more. We covered: Being collaborative at scale Avoiding adverse selection Getting to a yes instead of no Venture calendar audits Running the right strategy --- Timestamps: (0:00) Intro (0:47) The collaborative venture model (5:47) Adverse selection vs coverage (11:59) How to see a ton of companies (14:44) Getting to founders early (21:10) Helping teammates get to a yes (23:25) Why there aren’t more BoxGroups (27:57) What’s learnable about picking (31:46) Calendar auditing (34:39) Focusing on where you’re outlier (37:25) Depth vs breadth of network (41:03) The future of code (44:00) Brain computers --- More on Greg: https://www.boxgroup.com/ https://x.com/grosen More on Jack: https://www.altcap.com/ https://x.com/jaltma --- https://linktr.ee/uncappedpod Email: [email protected]

• 0:00Speaker 00

  You are better off just focusing on the thing that you are outlier in. And like, I think the biggest thing is just figuring out what that is.

• 0:06Speaker 01

  Which of the archetypes you are and then.

• 0:07Speaker 00

  Which of the archetypes, like if you have like hyper networker philosopher and specialist like there's probably more, but like of those three, like which one are you?

• 0:17Speaker 01

  Yeah.

• 0:17Speaker 00

  Where do you have an advantage? And then just focus all your energy and effort on being amazing there.

• 0:23Speaker 01

  All right, Greg, I am very happy to be here with you and I'm going to give a little disclaimer that I cajoled you into doing this and you normally wouldn't, but you're one of my best friends. And so you're doing this favor for coming on. So thanks for doing this. This is like your first podcast?

• 0:36Speaker 00

  This is my first podcast. I heard this is like a small boutique. No one really knows about it, right?

• 0:41Speaker 01

  It's a small little thing and you agreed to do it once and then you're.

• 0:44Speaker 00

  Never again.

• 0:44Speaker 01

  Never again.

• 0:45Speaker 00

  Yeah, don't invite me back.

• 0:46Speaker 01

  Okay, so I want to start with one of the things that you've taught me a lot about, which is the idea of collaborative venture. What's interesting about it to me is it used to be the case in venture that funds were smaller and collaborative was like the nature of how the whole world worked. And funds would like sort of like team up and like do rounds together. And then at some point in the last 10, 15 years, I think it swung to a place where funds got huge, people got sharp elbows, people wanted lots of ownership and whatever. And so a lot of that went away. And so in some sense that declined, but Box is like bigger than ever, doing more than ever. And so I wanted you to just start by hearing your overall view on this like collaborative venture model.

• 1:27Speaker 00

  Yeah, collaborative venture is hard. And almost by definition, it has to be anti-scale in certain dimensions, right? And so for us, like if we started leading seed rounds or series A rounds and certainly beyond that, we would lose our collaborative Switzerland nature.

• 1:47Speaker 01

  Yep.

• 1:48Speaker 00

  So instead, what we said is, okay, that is an easy way to scale capital is start leading rounds, writing larger and larger checks. And that's what others have done and why it's like hard to stay collaborative.

• 1:59Speaker 01

  That's like usually what happens, right?

• 2:01Speaker 00

  That's usually what happens, right? Everyone starts a new fund and or new firm and they say, we're going to be super collaborative. We're going to work together. And then all of a sudden they start. Leading one, two, three deals. And the second that happens, it's this binary moment.

• 2:14Speaker 01

  You either make it or you don't.

• 2:15Speaker 00

  Right, because if you're, if you're, if we're collaborative, right? We're a box group. And then let's say we make 80 investments a year. And let's say we only lead two or three seeds of that. Now, when I'm sharing a deal with you or anyone else in the market, you're going to be thinking in the back of your mind, is this fully being shared, all the deals? Or is it the one or two deals that they're leading and actually most excited about? I'm not getting shared, I'm not seeing. And so it's this binary moment that like the second you start leading those rounds, you just can't be Switzerland. And for us, we have been, you know, since the beginning of Box Group, you know, 15 plus years ago to now. It is like, we are not doing that. We are Switzerland. And, and, and that's, I think, what's enabled us to like.

• 3:00Speaker 01

  So just as like a little sidebar on that, does that mean no matter how excited you are to seed? You think it's amazing, crazy differentiated access? It's long-term not worth it for you to break the model and just like lead a seed?

• 3:11Speaker 00

  Correct. It's why we, like, our check size from, you know, the beginning to now has gone up. But it's basically scaled directly with like the inflation of rounds. And so, you know, seed rounds a decade ago might be like a fully loaded seed round might have been $2 million. Now that's $5 million. And so our check size started at, you know, way back in the day, 50, 75K. Now I would say it's like 750 to a million. And the reason we do that is that we want to make it as easy as possible for a founder to say yes to us. And so that means if we get to a founder first, we can commit a million dollars or 750K and be like, hey, you don't have a lead. We are aligned with you. You're raising three, four, or five million bucks. We're going to be your first yes. And let's go help you find an amazing lead.

• 3:56Speaker 00

  Or we could meet a founder and they'd be like, hey, we already have Jack Altman as a lead. And we're like, great. We can sort of round out a syndicate very quickly here and we can also get to a yes. And so the, the design for us is make it as easy as possible for a founder to say yes in whatever the round structure is. But it is not writing a three, $4 million lead check.

• 4:17Speaker 01

  Why do you think that more people haven't? Scaled it up because there's like a lot of funds out there that are 20 to 75 million that do, you know, 250K, 750K checks. But most don't scale it. Most basically get to this jumping point that you've talked about where you're either like, I gotta go lead rounds and like compete to lead seeds or not.

• 4:37Speaker 00

  Everyone wants to scale, like, anyone who's doing this is ambitious, right? And wants to get to scale. And so there's really only two paths that we've seen to like, scale the craft of what we are doing. One, scale AUM and then go further up the stack, start leading seeds and As and Bs, write larger checks. Or, and sort of the approach that we have taken, is do more deals, right? Scale velocity in terms of early deals with this collaborative Switzerland structure. And why do more people, you know, choose the, you know, former versus the latter? It's really hard, right?

• 5:16Speaker 00

  Like, if you took a look at any one of the people who work at Box Group's calendar, the sheer number of net new deals.

• 5:27Speaker 01

  Is ridiculous.

• 5:27Speaker 00

  Is ridiculous. And I think if you audited the average partner, or general partner, or managing partner of a lot of other firms, and you looked at how many net new deals they are meeting a week, I think you would be shocked.

• 5:40Speaker 01

  Yeah.

• 5:41Speaker 00

  Like, truly shocked.

• 5:42Speaker 01

  I want to come back to the calendar audit because that's like one of my favorite topics with you, and I think everyone gets this super wrong. But on the scaling, what is it like? I guess, first of all, I think implied in what you said is that one of the big deals here is not getting adverse selection, right? And like, I guess when you're thinking about scaling and what makes this work, to what extent do you think it's about you not getting adverse selection? And to what extent is it just about you seeing tons and tons of deals?

• 6:07Speaker 00

  Look, it's both, but it's primarily not getting adverse selection.

• 6:11Speaker 01

  That's like the big thing to solve.

• 6:12Speaker 00

  That is the big thing, right? Like, every strength has a corresponding weakness and vice versa. So, the knock on, that I hear from other people knocking on our model, is like, you need to own a ton of the companies that work in order to make venture work. And for us, what we would Say is that is true, but also you gotta be in the right companies and it's much harder to be in the right companies. And so what we do is we trade ultimate ownership percentage points for a more collaborative, flexible model that lets us just see better companies and hopefully get into more companies.

• 6:48Speaker 01

  And then the math basically is you gotta get into the $30 billion companies, not just the three, but by doing a bunch of companies, your odds of that go way up.

• 6:55Speaker 00

  Exactly. Like we are, you know, the, the other thing that I hear sort of as a knock against our model is like, well, this is an index, right? And, um, you know, again, we are holding such a high bar. We are like always asking ourselves, is this gonna be the most important company started over the next decade? But it's really hard. And so we take more shots on goal to try and get those outcomes. And then when we capture them, you know, it's our job, you know, to continue supporting them and piling money in.

• 7:23Speaker 01

  Gary Tan talked a lot about this. Like he said this on Twitter. I think we talked about it here that he was like, basically, if you look at all these funds that just do very big baskets of Y C, they outperform like ridiculously. And it's like, hard to believe that it could work like that, but it somehow seems to work like that.

• 7:38Speaker 00

  Yeah. I mean, I don't know the exact statistics. Um, and, you know, someone from Y C should correct me, but it was like, the class sizes were about 100. And I think it was like, you would get two to three billion dollar companies per batch, at least historically. And so let's call it whatever it is, one, two, three percent. Where else can you get a, you know, one in 100, two in 100, three in 100 chance of hitting, you know, a unicorn or a multi-billion dollar company? And that's what makes it so great because that pool is sort of a really elite pool.

• 8:12Speaker 01

  Yep.

• 8:13Speaker 00

  And that's sort of what we think about internally at Box Group. Now, we have to sift through a lot more than 100 deals to get there. But, but that's sort of the Box Group machine that we talk about in terms of trying to just see as many deals as possible.

• 8:27Speaker 01

  I think implied in your worldview about adverse selection is that at seed, it is somewhat knowable what's good. In other words, like you're saying, like you'd rather get to go alongside Sequoia than have to compete with Sequoia or whatever. There's like other views that like seed is like super unknown and nobody can tell. But I think your view is that like some, tell me this is a fair characterization, something like picking is actually, you know, most people can see a good thing when they see it. And you actually just have to like see the stuff and just like get in some way, somehow.

• 8:58Speaker 00

  Yes, I would say especially as venture shifted from a cottage industry to like, you know, a machine industry with tons and tons of capital and tons of firms. I think there's no such thing really as a proprietary deal. And if it is, it lasts for like literally a blink of an eye. Yeah. Like not even a day anymore. And so what that means is, is just like, it's the whole, you know, inner workings of venture is just way, way harder. Yeah, like competition has gone through the roof. And I think what's interesting to me is that most, most people, if not everyone, limited partners, general partners, employees, founders, I think everyone would agree that venture has gotten more competitive.

• 9:43Speaker 00

  Yeah. Has gotten way harder, right? And what's interesting then is that if you look at the math formula of venture, the like seeing, picking and winning a deal and then equals enterprise value output. Everyone says like, okay, those are the three things and venture has gotten harder. But if you don't change any of those variables, what are you accepting? You're accepting that returns are going to go down. Like you're just going to perform worse as a, as an investor. And so then what the counter is for most investors. We'll be better at picking. We're going to be better at picking. We're like, it's gotten harder and we're just like, oh. Think harder. Yeah, we're gonna, exactly. We're going to be thought boys. And in reality, like maybe I'm just not good enough at this job.

• 10:24Speaker 00

  I don't know how to be 50% better at picking, let alone 2X, let alone 10X. I just think it's like a fool's errand to think that you're magically going to be, become better pickers. And so for us, we've said, okay, venture has gotten harder. We don't know how to be better pickers. Let's take winning aside. We are going to be really focused on seeing. How do we, how do we like see more in order to?

• 10:51Speaker 00

  Like counteract how much more difficult it is. And then let's have a model back to like making it easy for founders, you know, to get to a yes to allow us to win more. And what I would call this is like, you know, Josh Koppelman came on your podcast and talked about the venture arrogance score. I would say this is our like venture humility score, which is we don't know how to have a basket of 20 companies in, you know, a fund or, you know, in a year and pick the most important ones, given how competitive and hard it is. So for us, we're saying we need to do more. We need to see more. We see on average about five to 6,000 qualified or referred opportunities a year.

• 11:34Speaker 00

  And then we at Box Group end up making about 70 to 80 investments a year. And like, that's our humility is like, if I could at the extreme, make one investment and I would know that that investment would be OpenAI, of course I would do that. I'd have the best performing venture fund of all time. We don't know. And so we take more shots on goal because it's much harder so that we can counteract how much more difficult and competitive venture has become.

• 11:59Speaker 01

  So then the game is like, see a ridiculous amount.

• 12:01Speaker 00

  Exactly.

• 12:02Speaker 01

  So how do you see a ridiculous amount? Like, what is the body of work that goes into saying, we can't get better picking, but we can see a bunch more stuff?

• 12:08Speaker 00

  I would break it up into inbound and outbound. Inbound is like the icing on the cake that, you know, tastes really great. It's awesome, but you can't just eat icing, right? You need the actual body of the cake and that is outbound. So we really focus on a bunch of great outbounding activities. That looks like running a ton of different events across the country. So intern events, events on campuses, and you know, engineering leaders. And again, these things aren't unique in terms of like, oh, we figured out unique alpha here. No one else is running events, but like, quality matters here, right? Again, back to like empty calories.

• 12:50Speaker 00

  It is very easy to do a bunch of events that return nothing because you didn't curate great people and do all the hard legwork. And I think, you know, our, our team does an exceptional. Job at running and doing these events. I think probably 25% maybe slightly more of deals is is from the sharing and sort of Switzerland work that we do with other investors and people. And why do they share openly with us? It's because we're not going to snipe a deal. We're not going to like compete with them for a lead position. And more so, they want to be preferred partners for, you know, when a company that we are fortunate to back in the pre-seed or seed in flex, we can, you know, say like, hey, we've loved working

• 13:33Speaker 00

  with Jack Altman, and so you should go take, you know,

• 13:37Speaker 01

  Donwell.

• 13:37Speaker 00

  Exactly, his money for the series A.

• 13:40Speaker 01

  You don't just have somebody, for the most part, who's just like reaching out to people. Like, it is, to some extent, this inbound outbound mix where you're kind of like warming up groups of people that you think are interesting and you're talking to your network. Or are you also like really reaching out and like, just like, you know, machine gun type ways?

• 13:55Speaker 00

  Like, we built some software to do like LinkedIn scraping when someone like changes their job title or whatever. It would be great if someone like posted on Twitter, hey, I'm amazing and awesome and I'm starting a new company. Like, that doesn't happen. And usually when that does, it's like way too late. There's already probably been two rounds into the business. And so for us, like, we really think about, like, we have the saying internally, like, if there's a deck, it's too late, right? It means that not only has there been like a team coalesced, but enough work has gone in to think to build an actual deck. And that means we're late to the party. And so so much of our outbound and work gets to like, before there's a company, there's a human being.

• 14:36Speaker 01

  Yeah.

• 14:37Speaker 00

  How do we get that human being oftentimes before they've even left the company? And that's what earns us the right to be that first check, that first yes.

• 14:44Speaker 01

  How do you do that part? Because this is one of the things I've always, and you invest mostly a lot earlier than I do, but, you know, I've always wondered, like, how do you get to people at some, in some reasonable way that will make sense at a top of funnel? Because in my mind, you need to be talking to like thousands and thousands of people that hopefully someone's going to leave at some point, at some way, like,

• 15:03Speaker 00

  Yeah.

• 15:03Speaker 01

  How do you do it?

• 15:04Speaker 00

  The, almost like tactics aside, getting to the mentality, because I think that's like most important here. Back to the calendar audit. There is no substitute for your time, right? And most people, especially check writers, do not spend time with someone who hasn't left their job yet, who may never start a company. That's primarily how every single one of us at Box Group spends our time, is trying to meet and curate those people before they've even left their job. And so step one is there's no substitute for putting in the time. And you have to be okay with taking a lot of really, I don't want to say bad meetings, but like bad meetings.

• 15:47Speaker 00

  And they can exist in different ways.

• 15:50Speaker 01

  It's not that it's bad people. It's that it's meetings with a director of engineering at a cool company. And you spend an hour with them and it turns out they just love being a director of engineering.

• 15:58Speaker 00

  Exactly. Like, you know, we have different sort of verbiage internally when we meet, unlike how we describe and talk about founders or companies. And like, sometimes you get, you know, someone who is an amazing big company person, who's not a founder. Sometimes you get an extremely technical founder who's not commercial. You get all these things. And by the way, you then have to line up those different archetypes with what are they working on? So like, you can't make these blanket statements. But more so, my point is when you go super early and nothing is formed, you cannot look for perfection. You cannot look for cohesion even in ideas, right? Like, oftentimes we've missed and passed on companies because we were like, we met this human, they were really interesting and amazing, but like, they were all over the place.

• 16:42Speaker 00

  They're talking about 10 different ideas. And it's like, that's a feature, not a bug. That means we're getting to them before they've actually figured out exactly what they want to work on. And ideas are not precious, you can work on anything. And so now we've sort of iterated away from that and said like, okay, that's actually a sign that we're doing our job and getting to there early.

• 17:02Speaker 01

  I guess it's also not just people leaving companies. You're probably getting people out of school and stuff like that too, which is another big body of work, I guess.

• 17:08Speaker 00

  Absolutely, absolutely. Like, the number, and look, I think that has increased dramatically with the rise of AI.

• 17:15Speaker 01

  Because it favors younger founders?

• 17:17Speaker 00

  It favors younger founders. Like, whenever you get new technology, right? Like it's always the younger generation that figures out the most native ways to use that technology.

• 17:27Speaker 01

  And even the old ways of building a company become less relevant, so the advantages of experience decline.

• 17:31Speaker 00

  Yeah, like the rules get rewritten. And I've like, you know, certainly felt that with like Cogen and other areas now. It just feels like every day, something new comes out, some new piece of technology is what makes the job so great. But like that like favors young folks. And I think like if you charted the number of like 18, 19 year olds that like we are funding now versus 10 years from now.

• 17:53Speaker 01

  Way up.

• 17:54Speaker 00

  Is, I mean, I don't know off the top of my head, but it's like at least one order, if not two orders of magnitude higher.

• 17:59Speaker 01

  Wow. Yeah. When you're bringing people in, or I don't know, to the extent you'd like bring them in for a partner meeting, but like.

• 18:07Speaker 00

  We don't do partner meetings.

• 18:08Speaker 01

  You don't do partner, yeah.

• 18:08Speaker 00

  I think partner meetings make no sense.

• 18:10Speaker 01

  Why?

• 18:10Speaker 00

  One, like a partner meeting is designed so that you get multiple like partners or decision makers to like give you an opinion on the person. And what we've found is one, very few founders like doing that.

• 18:28Speaker 01

  I hated it.

• 18:28Speaker 00

  You're just regurgitating the same thing. You're getting everyone's like.

• 18:31Speaker 01

  Super performative too.

• 18:32Speaker 00

  Exactly. But like that aside, which is like a huge waste of time on the on the founders part. For us at Box Group, it's like, we've found that like, there's three classes of deals that come in that we discuss, right? As a group. Obviously great, obviously bad, everything in the middle. The obviously greats are easy. It's like, all right, let's go, you know, find a way to invest. Those actually, interestingly, as a basket, perform worse. They're higher priced. Usually they're more like likely to get acquisition offers. Obviously bad, we don't spend any time on. And then all the alphas in the middle bucket.

• 19:08Speaker 01

  But you do the obviously greats anyway?

• 19:10Speaker 00

  We do the obviously greats anyways.

• 19:11Speaker 01

  Looking back at your biggest hits, have any of them started in that basket and then in fact been obviously great? Or have most of your, or have all of your really good stuff?

• 19:19Speaker 00

  I don't want to say all, but like most.

• 19:21Speaker 01

  Wow.

• 19:21Speaker 00

  And it's all for different reasons, right? But, but, but this sort of gets to my point, which is the. Sort of middle is where all the alpha is. And because it's so early, like oftentimes, the deals, deals that are there are pre-product, oftentimes pre-team, like there's nothing. And more often than not, we are discussing those deals as just like, we're talking about the human being for 15 minutes, right? And like a little bit, maybe two minutes on what exactly they're working on.

• 19:55Speaker 01

  So what are you talking about with the, when you're, when you're having a debate like that? What are the things you're debating? Like, what are you talking about?

• 20:01Speaker 00

  We're basically looking at, I hate the term spikiness because it's so annoying, but like maybe said a different ways, like we're, we see, again, 5, 6,000 opportunities a year. What is outlier? What is standing out? What is just different? What is making us likely to want to go quit our jobs where we get a portfolio and go work for them, right? When we see all these really interesting companies. And that's different. Sometimes that can look like they're talent magnets. Sometimes it's like we're talking to an 18-year-old and it feels like you're talking to a 30-year-old. And they're like mature beyond their years. Sometimes it's amazing technical prowess or expertise. Sometimes it's like jump-off-the-page commerciality, right?

• 20:40Speaker 00

  But, but, but I guess my point is, almost always, in that bucket, it is not obvious. It is like, we're talking about it in an obvious way. But like, these are decisions that are so on the margin. And it's like, if you wake up the wrong way and you're kind of in a grumpy mood, and that causes you to pass on the one, you know, great, like-

• 21:02Speaker 01

  It's so expensive.

• 21:03Speaker 00

  Hundred-billion-dollar, one-trillion-dollar company that gets created either that year or that decade, that is extremely costly. And so instead, what we've found is that we should design a model that, like, we are trying as a team, instead of sort of poking holes and trying to figure out all the ways that it can't work, which like, I could do, by the way. I could have every single company that comes in, even the good ones, I could find a million reasons why it's going to fail and not work, and be very negative and try and talk all my partners and team out of it. We take the opposite view, which is, we have to help- The sort of person that's dreaming with the founder or squinting is sort of the term we use internally to sort of get to their own version of yes.

• 21:46Speaker 00

  And so what you hear is that the sort of individual stakeholder that's meeting the company will sort of talk about it as the group. And then all of us, instead of poking holes, and it's not that we're not critical, we try and pull on the thread on, hey, like you mentioned this language. That seems pretty outlier to me. Like, can you expand on that, on like what you saw in this founder? And so we're all trying to help them basically get to a yes in our model, which is very, very unique. And then we are, you know, single trigger.

• 22:17Speaker 01

  I feel like that's your whole orientation. Like even when I call you, which I do all the time about like a deal, you're often trying to help me get to a yes when like almost the whole industry is trying to get you to a no in some weird way.

• 22:29Speaker 00

  Yeah, and this is the, you know, cost of omission versus, you know, commission and

• 22:33Speaker 01

  It goes back to you doing a lot of companies per fund.

• 22:35Speaker 00

  Exactly. It's like you just, the number one thing that matters is just being in the right companies. And, and you know, and I remember, you know, I won't mention the name of the company, but like you called me on a, on a deal that you were looking at. And the first conversation, I was like, here's the critical pieces of the market and product and like the competitive landscape. And sort of, I was like, you should think about that. And then I remember you called me again later that week. And intuitively, I didn't realize I was doing it. We were just debriefing earlier than this. But then I swapped in that second meeting to like, oh, you now are seeing something, you're squinting and seeing something that's not obvious in the market. Let me like help pull on those threads to help you get to a yes. And you ended up investing in the company.

• 23:21Speaker 00

  And, and I think that's what we aim to do as an entire team across the board, not just partners, but like a associate or analyst who just joins. It's like all, everyone, that's the mission and ethos of like someone at Boxer.

• 23:35Speaker 01

  So you've described now like a few attributes of doing like super early stage investing, which include like a big basket, an orientation of trying to get to a yes, not trying to get to a no.

• 23:45Speaker 00

  Yep.

• 23:45Speaker 01

  Looking for people, not ideas. All of those things are like extremely different than everything about how like a big firm operates. Like those are like opposites on all dimensions.

• 23:55Speaker 00

  Yep.

• 23:55Speaker 01

  I'm curious how you think about like a big firm trying to do this like, I don't know if you want to call it pre-seed or first round or whatever, but like they're just such different sets of work. Everything about it's opposite.

• 24:09Speaker 00

  Yeah, and it's, look, it's why I always ask myself like, why are there not ventures so competitive? Why are there not more box groups? And it's so counter to how so many people run their business and do their, you know, work. And I think, you know, for us, it's just the only thing we know. And I come back to like, if, if your ultimate goal and orientation is not, um, how do we maximize ownership? And then hopefully we are in some really good companies there. And instead say, how do we, if we look at all the good companies that get started in a year, we think about how many of those did we see?

• 24:51Speaker 00

  And then how many of those did we like actually get right versus wrong?

• 24:56Speaker 01

  Yep.

• 24:56Speaker 00

  And the way that we look at it, and I think, you know, my partner David Tisch, who started the firm owns a lot of, like deserves a lot of credit here, which is like, the failure mode is we didn't see it. If we made the wrong decision, okay, let's deep like figure out and debug that. But what is unacceptable is like, we didn't see this deal. And if we didn't see it, we got to figure out why. And, and what I would say is like, most people I think would look at venture and say the most important thing out of the C pick win is pick.

• 25:25Speaker 01

  Yeah.

• 25:26Speaker 00

  And for us, the ethos of box group is like, we don't earn the right to pick or win unless we see it.

• 25:30Speaker 01

  Yep.

• 25:30Speaker 00

  And by the way, seeing it late versus early can be the difference of winning or losing.

• 25:34Speaker 01

  Mm hmm.

• 25:35Speaker 00

  And so we are obsessed as an organization on seeing everything as early as possible.

• 25:41Speaker 01

  You mentioned the single trigger thing, which I think is, I don't know if it's completely rare, but it's less common than trying to get to some amount of group consensus.

• 25:49Speaker 00

  Yep.

• 25:49Speaker 01

  How structurally important is that?

• 25:51Speaker 00

  Incredibly consensus driven decision making at seed doesn't work, in my opinion. Back to the point on it will just shift the level of risk that you're taking, right? If we had to vote as a team on every single deal, one, we would need to ask founders to come in and pitch everyone. By the way, our model then would work, it would be too much time sink. What are we going to have like a full day of like the, you know, 12 or 15 companies that are interesting that week? Coming in and pitching the partnership, it would break down. But more so like it's again, back to that point, it's usually just one person who is like, or two people or like in a pod. We usually operate in pods, you know, our team size is like 11 investors, 10 investors. And, um, you know, we operate in pods of like, you know, two or three usually.

• 26:35Speaker 00

  And it's like, that's, it's usually even one person within the pod that's like getting really excited.

• 26:41Speaker 01

  It's actually double because you've got both single trigger, but you also have a culture where people are trying to get you to a yes. Versus, you know, you have got consensus and it's like anybody's trying to block. So you've got both things, which makes getting a yes, obviously like, you know, more frequent.

• 26:55Speaker 00

  Yeah. And I think because we are all obsessed with seeing deals, there isn't this concept of senior partners' time is too precious. Only loop them into a good deal, right? Like something that we try and untrain sometimes is like, you know, we'll hire someone new. They're like, starting as an analyst or an associate. They'll be like, oh, like, I thought this was an interesting deal. I'd love to loop, you know, someone in to get another, you know, set of eyes. And the second meeting, for whatever reason, is not good, right? And then they'll apologize. They're like, ah, I'm sorry. Like, you know, it didn't, like your line of questioning, you know, made me realize that it's not a good fit. And it's like, never apologize. None of us are too busy to take the meetings and like, it's your job to go and try and surface those interesting ones.

• 27:39Speaker 00

  Now, you can't always surface bad one and that deal taste gets honed. But like, there is this culture of no one's time is more valuable than anyone else's time. And like, we are trying to see. You know, deals as early as possible, which means we're gonna take a lot of bad meetings, and that's okay.

• 27:56Speaker 01

  Yep, yep. You've talked about how picking is like, not the thing.

• 28:00Speaker 00

  Well, it is.

• 28:02Speaker 01

  Well, I was gonna say empirically, you're a good picker. Like, you just, you just have been. And I guess my question is, is there anything describable about picking people or spotting, you know, I'm, we won't use spiky, we won't use taste, but is there anything about what goes into seeing something special about a person early that is describable? Or is it only learnable through lots of time? Or is it just not a learnable thing?

• 28:26Speaker 00

  I hate giving like, half answer, but the, the truth is, it's like, half and half. Half of it is just, you can't teach taste. Like, certain people just like, have inherent deal taste. And what is that? It's like, the human taste. It's like, who do I like spending time with? Who, like, do I think can just like, do the impossible and build?

• 28:43Speaker 01

  I've actually always thought about that, that like, people just like, have their preferences about who they like. For some people, that set of preferences is just gonna overlap with super talented people, and for some, it's just not.

• 28:51Speaker 00

  Talent clusters, by the way, like, always. And so, like, if you're a high-performing human, like, you are gonna spend time with other high-performing humans. So I think like, there's a natural clustering there, that like, you just can't, you can't teach.

• 29:05Speaker 01

  Yep.

• 29:05Speaker 00

  The other is like, you can get reps, you can practice, right? Like, you know, we, we both love playing golf. There are certain natural athletes, you know, in a big way. But, there's no substitute for just lots and lots and lots of reps. Boxer is like, I don't know how else you can get as many reps of seeing companies and like, across all different stages and sectors, we're generalists, we're not specialists. That's, I think, the other half of the equation.

• 29:33Speaker 01

  You mentioned something to me that I, it really stuck with me, which is, you said, basically, if you looked back, anytime you took a third meeting with a founder, you should have done it.

• 29:42Speaker 00

  Yeah.

• 29:42Speaker 01

  And I thought that was like, a really interesting tidbit, that basically, like, one of the ways to get better is to simply be more observant of your own intuitions, and the act of you continuing to spend time with somebody, even if you're not sure, is your intuition trying to tell you something.

• 29:57Speaker 00

  We're always updating our, we don't have like, any rules, but like, We have these like de facto sort of learnings and we're always updating them, right? Back to the like, you know, okay, we did a good job. We saw, you know, a bunch of interesting companies. We passed. Why? And a big basket of, you know, one of them was more often than not, I would say the average deal is two meetings, right? In this market. First meeting, intro, we're kind of figuring it out. Second meeting, both like go deeper, maybe have someone else come in and also sell the founder, spend, spend a bunch of time and like then invest, right? So that's like the average flow. There was a subset of deals where we were taking three, four, five, six meetings. And why are we doing it? It's because someone again was seeing something

• 30:42Speaker 00

  and was almost taking more and more meetings, trying to make it obvious one way or the other. And it's like, it's not obvious. It's just you gotta, you gotta believe or not and just do the deal. And so now it's like, if we're gonna take a third meeting, just do the deal.

• 30:55Speaker 01

  Yeah.

• 30:56Speaker 00

  So that's like one. Another one that sounds like obvious is that, you know, sometimes in really competitive processes, we, um, you know, will meet the, you know, CTO or co-founder, you know, in the first meeting instead of like the CEO, um, or we bump into someone and it's obvious like always meet the CEO. Or sometimes the CEO, it's like two sort of equal co-founders, just like always meet the, you know, one CEO. And then I would say the last one that we've really thought about, and I think like, you know, the mistake that we made and then how we corrected it was sort of not giving enough surface area to like, ideas aren't precious and like, we don't want to invest hoping for a pivot, but if the human's good enough, just like, don't overthink the market. Like either we're gonna be wrong on the market because it's so

• 31:40Speaker 00

  early, or the founders are gonna change and evolve. And like, you know, just don't overthink that.

• 31:46Speaker 01

  I want to talk about the calendar audit stuff. You and I have talked a ton about how like, people spend time.

• 31:50Speaker 00

  In very dumb ways.

• 31:51Speaker 01

  In very dumb ways.

• 31:53Speaker 00

  Lot of empty calories.

• 31:54Speaker 01

  It's funny. I think I also, you know, when I was running a company, I also spent time in dumb ways, but less. And I think in venture, it's really possible to spend time in dumb ways because everything is amorphous. You don't know what's actually gonna lead to what. And so it leads to, I think, like, I think if you compared calendars across venture capitalists, you would see like a ridiculously different pie chart of how people are spending time.

• 32:16Speaker 00

  Absolutely. And look, like, everyone's running a different strategy, right? And so I also understand that, like, for you and the sort of craft that you're doing in, like, leading Series As and being an amazing board member, like, that is going to shift the amount of time that you're going to spend and take on, like, completely net new meetings. However, you are only as good, and this was like drilled into me when I was at Benchmark, you're only as good as your last deal. Like, you have it really hard where you have to both create incredible founder MPS as like a great board member, which is also anti-scale. But then you have to be always spending time on finding that the next great company.

• 32:55Speaker 01

  Yep.

• 32:56Speaker 00

  And what I find is that most people are not honest with themselves and they hide behind that. They're like, "Well, I'm on 10 boards, so like, I gotta spend all my time doing boardwork." And I'm like, "Okay, but like, why are you spending, you know, all this time networking with like other investors?"

• 33:12Speaker 01

  Totally.

• 33:13Speaker 00

  "Why don't you go meet founders?"

• 33:14Speaker 01

  Yeah.

• 33:15Speaker 00

  Or like, "Why are you at this like event that's all VCs and no founders?"

• 33:18Speaker 01

  I think that stuff's a massive waste of time.

• 33:20Speaker 00

  Massive waste of time. And, and people think they're so busy, but like, there's always more hours in the day.

• 33:24Speaker 01

  And it feels good. You feel like you're out and about. You think you're doing stuff, you're meeting people, and what did you do?

• 33:29Speaker 00

  Yeah, yeah, yeah. And so...

• 33:31Speaker 01

  What are the other big time wastes? Like, what, like, if you had to like name the top time oysters? Like, I think networking events has gotta be...

• 33:37Speaker 00

  VC networking is like, gotta be number one.

• 33:40Speaker 01

  Yeah.

• 33:40Speaker 00

  Basically, anything that isn't, if, again, you're, you're in the boardroom doing boardwork, like, if you're not helping your portfolio, you should be spending all your time trying to meet and find new companies, but again, really individuals and great, like, interesting pockets of people.

• 33:59Speaker 01

  Yeah.

• 33:59Speaker 00

  Like, if you want to find alpha, you have to be people-focused, not company-focused. There are many ways to do that. You know, there's, like, we do sort of a hyper-networked approach at doing it. There's sort of a philosopher tweeting, media, very loud out-there approach. I also think there's being a specialist, right?

• 34:20Speaker 01

  Yeah.

• 34:22Speaker 00

  You know, this is the one area that basically I'm gonna know more than anyone else.

• 34:25Speaker 01

  Yeah.

• 34:25Speaker 00

  As like three different archetypes.

• 34:27Speaker 01

  So it's like, so there's like, there's like networks see everything. There's like, you know, media thought boy, like.

• 34:33Speaker 00

  Is that you?

• 34:33Speaker 01

  Pouring out for me. So that sucks. Yeah. And then there's like a specialist.

• 34:36Speaker 00

  Yeah.

• 34:36Speaker 01

  Those are like the three ways you can go about it.

• 34:38Speaker 00

  Yep.

• 34:39Speaker 01

  What can be a big problem in venture and just in general is when you start trying to get caught up in other people's games, you like, don't just make like no progress. You like go backwards.

• 34:48Speaker 00

  Yeah. You like, you spiral and waste a bunch of calories. And what's most dangerous is that it can, look, the tactics can be the same, um, but the quality can be very different. Right? And so like, again, I don't, I don't tweet. I don't do podcasts.

• 35:04Speaker 01

  You do this podcast?

• 35:05Speaker 00

  I do this podcast.

• 35:06Speaker 01

  That's right.

• 35:06Speaker 00

  Many other people have tried to create.

• 35:08Speaker 01

  We just do this once a quarter and that's all you need to do.

• 35:10Speaker 00

  Um, exactly. But like, many people have tried to create a bunch of other podcasts and like, you know, you came out and did an exceptional job and of quality. And now this is, you know, probably going to hopefully lead to a bunch of really great, interesting companies coming your way. If I tried to do the exact same thing, it wouldn't work. Um, and like the time spent and all of that. But it's, it's alluring to like, oh.

• 35:33Speaker 01

  Totally.

• 35:34Speaker 00

  I could go and have the, and then everyone's gonna come inbound to me and it's gonna be so great.

• 35:37Speaker 01

  Or he's like, I mean, this is gonna come off as a joke. So sorry, Sam, but like, you know, our friend Sam is like an incredible, he's a good at all thing, but he's really good security investor.

• 35:45Speaker 00

  Really good at cyber.

• 35:46Speaker 01

  Really good at cyber. Good at other stuff too, but particularly good at cyber. And you know, if we try, like if I tried to go be a specialist in something, I'm gonna like fall over. It's just not gonna work.

• 35:56Speaker 00

  Yeah, exactly. Like he should be better at that one thing. But the trap that I think, um, investors and especially like early investors in their career fall into is like, you know, the thing that you're good at. And then you're like, well, this other thing over here.

• 36:09Speaker 01

  I see people doing well with that.

• 36:11Speaker 00

  People doing really well. But the trap is like, you actually don't know how hard it is.

• 36:14Speaker 01

  Yeah.

• 36:14Speaker 00

  Right? Um, and it's like, you are, you are better off just focusing on the thing that you are outlier in. And like, I think the biggest thing is just figuring out what that is.

• 36:22Speaker 01

  Which are the archetypes you are and then get great out of it.

• 36:24Speaker 00

  The archetypes, like if you have like hyper networker philosopher, uh, and specialist, like there's probably more, but like of those three. Like which one are you?

• 36:33Speaker 01

  Yeah.

• 36:33Speaker 00

  Where do you have an advantage? And then just focus all your energy and effort on being amazing there.

• 36:38Speaker 01

  And there's probably some people who are like a major in one and a minor in others. But like even that, you need to like know what that is.

• 36:43Speaker 00

  Yes. Yeah. And back to like the calendar, like then audit your time and go back and look and be like, how many empty calories did I waste doing this thing? And like, even if the inputs are good, if the outputs aren't there, right? Then like.

• 36:55Speaker 01

  It also, it kind of goes back to your like inbound outbound thing where like inbounds icing. It's so easy to just get lost in, you know, just inbound requests. Will you do this thing or that thing or come meet this person? And, you know, one of the things that always shocks me is you could just like save 90 minutes by just, you know, saying a polite no, and then it's over. And people all the time just do stuff because they're like, eh, I don't have something else planned. So I can just do it.

• 37:20Speaker 00

  Say no all the time.

• 37:21Speaker 01

  Yeah.

• 37:22Speaker 00

  Like to coming back on this podcast.

• 37:23Speaker 01

  Yeah, that's right. This is the last, first and last. One more question and then I want to wrap with a couple of your thoughts on just specific areas of tech. As you reflect over your last sort of 10 years of venture, I'm curious how you've thought about like deep relationships, broad relationships, the sort of like tight knit versus professionalized nature of venture. Like what, what do you see over your sort of time doing this?

• 37:46Speaker 00

  What was really interesting is I, I feel like, like a decade ago, venture had clicks. Like venture both had clicks and founders had less power than they have now. And so what that meant is like, you know, VC would invest at like the series A and then go tell their, you know, founder like, hey, here's who's leading your B. Not always like, and I'm exaggerating a little bit, but like it was like that dynamic. And then what ended up happening is obviously founders got a lot more power, which is awesome. But then almost the playing field of, of like leads and new firms and everything, it, it kind of like spread out.

• 38:20Speaker 01

  Yeah.

• 38:20Speaker 00

  Right? Became more democratic. But that also meant that you were getting more sort of random people and sort of more junior people doing deals and, and sort of investing in companies.

• 38:31Speaker 01

  It was a bit of a mess. You're talking like early Zerp.

• 38:34Speaker 00

  Yeah, exactly. And, and now what's interesting is I feel like we are going back to like. Not clicks, but almost like who are the preferred partners? And what's cool is that founder sort of powers continued to go up.

• 38:49Speaker 01

  Yep.

• 38:49Speaker 00

  So what that means is, is that, and we see this all the time in, you know, people that we like working with, is we invest early, we sort of build founder trust and relationship. And then we say like, look, we're not, you know, leading your series A, who's like, here are series A investors who we really respect, who have done exceptionally well by like other portfolio companies and founders in our portfolio. They are our high rec. And what I would say is like, when I started in venture, I always thought it was like bad to play favorites. And now it's like, I'm absolutely playing favorites on the recommendation. And then ultimately it's on the founder to go and take that recommendation. And if we have high trust, like hopefully it means a lot.

• 39:34Speaker 00

  And then ultimately they're going to pick and choose what's best for their business.

• 39:37Speaker 01

  It's a very interesting observation because it's like a similar net effect, but for opposite reasons and with an opposite power balance. Where like 10, 15, 20 years ago, it was like VCs had so much control that they could just say, this is the round you're doing. And then it became, you know, founder power grew, venture capital dollars grew, and it became a bit of a mess. And what you're describing now is more a world where part of the job of the existing investors is to help the founder who does have all these cards circumvent a lot of the messiness there and say, you know, these processes are really quick. We know people that are great and like here's options presented to you.

• 40:11Speaker 00

  Yeah, and there's so many ways that you can get sucked into a process that's either annoying, a waste of time, or like worse, like a partner that has a great brand, but then in reality is like annoying to you.

• 40:23Speaker 01

  Yep.

• 40:24Speaker 00

  Right? And I think that's part of venture funds are sort of becoming hyperscalers in their own way and amassing huge billions and billions of dollars, building platform teams, all this different stuff. And it's not saying that that's good or bad, but it just means that there's more noise.

• 40:37Speaker 01

  Yeah.

• 40:38Speaker 00

  And so for like us, one of the things, you know, I always say like VCs love pitching magical value add. Like we pitched zero, like. Founders are the reason why companies work, not investors. But the one thing that I think we can do a really great job at is say, hey, across our 500 company-acted portfolio, here are the partners who have done exceptionally well by the founders that we've worked with.

• 41:02Speaker 01

  Yeah. All right. I'm gonna hit you with two quick future guesses and I'm gonna let you go.

• 41:06Speaker 00

  Okay.

• 41:06Speaker 01

  And then you can never go on a podcast again.

• 41:08Speaker 00

  Great.

• 41:08Speaker 01

  One is Future of Code. I just posted this podcast with Dwarash. He dropped a really interesting stat, which I think has also been shared by one of the Colsons. Guillermo has talked about this, but basically it was about these tools are not yet that productive. Like, obviously, people are finding ways to be very productive, but they're not yet. You have made amazing early investments in Cursor and Warp, which are awesome, sort of, leading products here.

• 41:37Speaker 00

  Yeah.

• 41:37Speaker 01

  So you know more than average about the space. Where are we right now and where do you think this AI code gen stuff is going?

• 41:43Speaker 00

  I don't think anyone knows, and it's changing so quickly, so I'll caveat by, like, no one knows. The thing that I feel strongly at is that whenever there's a new platform shift, and I started my career with the rise of mobile, right? And I remember when mobile first came out, people were like, taking HTML web views and taking their mobile website and saying, there we go, there's our mobile app, right? And that was the, you know, I always like talking to, showing how old I am, talking to young AI founders and being like, do you know when the App Store came out, what the number one app in the App Store was for like, the first two years? Do you know?

• 42:19Speaker 01

  I remember like, the beer drinking game. Was it that?

• 42:21Speaker 00

  No, that was, that was quick.

• 42:24Speaker 01

  For like, a day.

• 42:25Speaker 00

  It was Pandora, right? And the reason is, like, that's what us humans do. We're like, ah, like, an iPhone comes out, it's a supercomputer in your pocket, what's the most important thing? It's like, taking the radio that's in our car, putting it on our mobile device, right? And what happens is, it takes like, two, three, four years for people to actually understand the primitives that end up coming out of mobile, right? And then we get Uber, the remote control for the real world, and Instagram, and Snapchat, and Discord, and all these like, amazing, great, native expressions of mobile. That exact thing is happening, bringing it back to AI and code gen. But it's really hard, right? It's the reason why I think Cursor's done so well is that they took the IDE form factor

• 43:07Speaker 00

  that everyone knew and then built on it and made it a lot better. Yep. What Warp is doing with like the IDE and the Adeptus Development Environment is like, hey, this terminal and Claude Code and all these other components, we can just tell a computer what to do and maybe the right native expression is not opening up a code file and then hitting Tab 100 times because you shouldn't even be looking at an individual code file. You should just be telling the computer, you know, or machine what you want it to do and then sort of auditing all the like thought and steps throughout it, throughout the process. And then eventually, maybe you don't even need to do that. I'm always asking myself whenever a company is pitching us, what is the most native AI version of this?

• 43:50Speaker 00

  I don't know the answer, otherwise I would start the company myself, but like that's our orientation and that's been some of the, it's been cool to see the CodeGen, you know, companies that we've invested in take different steps at doing that. All right, one more just for fun. Neuralink, brain computers, what do you think about that? We were talking about this earlier where I was like, you know what? I don't know if it's harder or easier, but like there's so much stuff that's happening in like robotics and, you know, humanoids and all this amazing things. Maybe what's actually more fun and interesting is just like if we find a way to have input-output to our human brain, like then you just hook us up and, you know, it's like Matrix 2.0 and that actually may be easier than having like a Westworld environment.

• 44:34Speaker 00

  God, that's good. I don't know, it's like a little futuristic, but I'm surprised that there's not more people that are like focused on instead of sort of changing and almost terraforming the Earth and world as it exists right now. Just make our, yeah. Just like. Just put in the brain. Literally just simulate a much better one for ourselves and really what is the difference there? It's a little futuristic, but it's just a fun thought. That's fun. Because it may be, ultimately, it's like, what is the path of least resistance? I'm talking way out of my scope of knowledge. Yeah. But the path of least resistance, rather than that utopia on Earth, might just be we're all like hooked up to the matrix and, you know, we have perfect input-output to

• 45:15Speaker 00

  our brain and that may actually be easier than sort of the utopia with robots running around and humanoids and all that stuff. I don't know. Sounds good, God willing. Greg, you're the best. Thanks for doing this. Thanks.