Scott Kupor on A16Z Origin Story, White House Role and Office of Personnel Management

I'm a Ben and Mark lifer. Basically, the two of them had this vision that the venture industry had changed, would change. Um, you know, turns out 16 and a half, 17 years later, uh, they were pretty much right on spec for that.

So you've probably never heard of it, which is not surprising. We are basically kind of the talent and the HR policy organization for the civilian population in government. It is the biggest scale HR job, I guess, is the way to think about it.

Uh, I'm glad to be here. Thanks for having me.

Yes.

Yeah.

That's correct.

I did. I did. I'm a Ben and Mark lifer, basically.

A glutton for punishment is maybe the best way to describe it. [laughter]

Yeah. So, um, the history is it starts a little bit before, which is, uh, we were all at, at the time it was called Opsware. It started as LoudCloud and became Opsware. We sold Opsware in 2007 to Hewlett Packard, uh, and a bunch of us had some, you know, indentured servitude obligations to Hewlett Packard, uh, that we had to stick around for. And then, uh, but what happened was, you know, both for Mark and Ben, you know, they had started doing angel investing on their own probably going back to like '05 or something.

And, uh, they did a bunch of stuff and about, you know, 2008 they kind of said, hey, this is interesting stuff, but we think we could have a much bigger impact on the ecosystem if we raised a fund instead of just, you know, writing $50, $100,000 checks as individual angel investors. Um, so, uh, they were kind enough to reach out to me in the summer of '08 and they said, 'Hey, we're thinking about this thing about starting a fund.' And I was like, 'Uh, okay, that sounds interesting, but you got to remember this is like summer of '08.' So, uh, Bear Stearns had gone like bankrupt in like March of '08. We hadn't yet quite gotten to the depths of the financial crisis. Um, in September, of course, Lehman will basically end up going out of business. Uh,

And so I was like, 'Guys, like, this probably isn't the best time in the world to actually start a new fund.' And they're like, 'Oh, don't worry about it.' Remember, uh, in the history, the reason why they said don't worry about it is because we had taken LoudCloud public in literally, you know, kind of 2001. There was us and like one other company, Storage Networks actually was the name of the company that went public. So they're like, look, these things are fine, it'll all change. So anyways, that was kind of how it started. Um, I hemmed and hawed longer than I should have. Mark and Ben like to joke that, you know, had I agreed the first time they asked me, it would have been, you know, Horowitz and Andreessen Kupor or some version of that. I have no idea if that's all revisionist history or not, but I, uh, I squandered my opportunity to truly be a founder because, uh, you know, uh, I was too scared to basically jump ship in the middle of the global financial crisis. But, uh, I did start, you know, um, kind of when we, uh, you know, closed the first fund, which was kind of beginning of 2009. So look, the basic origin story was this. Look, the two of them had this vision that the venture industry had changed and would change, that obviously with cloud computing, with kind of, you know, the opportunities for people to be able to not have to use all their venture money to go buy servers and buy networking equipment, all this stuff, but you could rent this. It would just dramatically change the size and the scope and the opportunity for venture. And, um, you know, turns out 16 and a half, 17 years later, uh, they were pretty much right on spec for that.

Yes.

So yeah, so, you know, some of your audience may not know Michael Ovitz from his like super amazing glory days, but, you know, he starts what becomes CAA in 1975. And, uh, at the time his big innovation in the industry, which is a lot of, you know, as you're right, a lot of what inspired Mark and Ben was this idea that in the old, you know, Hollywood talent agency business, it kind of looked like the old venture capital business, which is you had like, you know, you were a GP in the VC world or you were a talent agent in the agency world and you kind of ran your own book of business. You know, you had your clients, you introduced them to opportunities in TV and newspaper or, you know, books or all kinds of other stuff, but it was kind of very much a like one-person show essentially. Like even though you were part of a firm, you really didn't have a concept of being part of a, you know, kind of consolidated effort. Uh, and, you know, what he brought to CAA was, look, like we're going to basically break that model and instead of having like the agent do everything, the agent is going to principally be the relationship management, you know, kind of arm of the organization and then we're going to build out expertise in all these other areas. So we're going to have somebody who just only does like television contracts and somebody who does radio and somebody who does movies so that when you come in as a talent, you know, in, you know, 1990 when by then they were basically, they had like 90% market share in the business, they would all sit around a table and, you know, the GP equivalent, the talent adviser would say, great, let me introduce you to my, you know, new client here and you would see all the people around the table who'd say, great, like I'm the TV guy, I'm going to show you what to do. And so the theory was to essentially decompose the job and then build out like, you know, expertise in different areas. And that was the basic kind of model that we did at Andreessen Horowitz was the same thing, which is, look, the GP obviously ultimately is the relationship manager and obviously the board member who kind of works with the, um, individual, but the GP doesn't have to know every single buyer at every tech company to introduce the company to. They don't need to know every CFO in the valley to, you know, bring them into their companies. The theory was build out these areas of specialized expertise so that when we present to the entrepreneur, we present as a unified firm.

And, uh, the canonical VC model was we basically said we were trying to institutionalize John Doerr again for, you know, you know this, but for your, you know, some of your listeners, John Doerr was like the canonical GP at Kleiner Perkins who basically, in my, from my perspective, made Kleiner Perkins. And John was, he was the everyman basically. So John knew, when you met with John, he'd be like, great, like I know all these people in politics, I know all these people, and he would be the connector. And so our theory of the firm was, could we institutionalize that? And so there wasn't a single thread of John Doerr, but there was a firm that you could interact with that basically could bring to bear all those relationships.

Well, the GP, yeah, look, the fundamental constraint in the venture business, as you know, look, is yes, the GP job is fundamentally limited by like how many board seats can you do. Now there is a perennial argument, which is how many can you do? So there's supposedly a guy at NEA who has like 40 boards. Uh, I won't name him, but, uh, who like, you know, appears to be breaking the mold, but right.

But yeah, so like we couldn't solve that problem, but what we could do is, look, if a typical GP could only do eight boards because they also had to spend time, you know, meeting the buyers at, you know, Home Depot and places like that. Instead, we could say, look, you don't need to know the buyers at Home Depot. We'll staff an organization to be able to do that. And so, number one, we could create greater scalability for the general partner. And so, I think if you look at our partners, some of them have like a lot of, you know, like even more, but let's call it maybe you could do 12 to 15 boards instead of eight or 10 because like we've given you scale on your time.

Um, but, uh, but yeah, look, that's the fundamental limiter. The other fundamental limiter, at least I think, and this is where I think Mark and Ben have brilliantly solved this problem, is at some point in time these partnerships get so big that it's very hard to make decisions because you got too many people sitting around a table arguing about stuff where, you know, some people actually truly have all the domain expertise and others don't. And not because they're bad people, it's just, look, if I'm Martin Casado and all I do is AI infrastructure all day long, of course he knows like a thousand times more about that than, you know, than I do or even someone like Ben does, right?

And so the other kind of, I think, innovation that these guys did in addition to Michael's innovation was figure out how to scale a venture capital firm. And like, again, you can kind of modularly scale at least the theoretical board capacity of a GP, but even that runs out. The real question is, can you create an organizational structure, which they did, where you basically subdivide into subject matter expertise at funds so that, you know, we've got a bunch of different funds, most of them have somewhere between three and maybe six general partners, uh, and recognize that, look, as brilliant as Chris Dixon is in crypto, he doesn't need to sit in like the bio and health pitches basically and, you know, try to add value to that because like it's just not his area of expertise. So that's, I think, the other like innovation that I think people forget about is just this idea of kind of scale within a venture firm that enables you to create a well-functioning like big firm like A16Z as opposed to like, uh, the centralization model, which I think is how most people have tried to kind of build scale. And in our view, like centralization just doesn't work in this industry because the domains are just too deep.

Yeah. So I would think about it as like there's two kind of vectors of centralization. So one is, look, the thing we don't want to centralize is exactly where we started, which is, look, you don't want to centralize decision-making. You want that pushed down as far down the organization as possible to the subject matter experts. Then there's like, okay, we started with go-to-market as a centralized organization, but over time as we added bio or crypto or stuff like that, we're like, wait a second, what an entrepreneur needs in go-to-market in crypto is different than what a bio person needs. And so over time we decentralize those. So today there is a go-to-market team that lives in the crypto fund with Chris Dixon and there is a bio go-to-market team. Um, so we kind of started to disaggregate things that initially were kind of, uh, centralized, so talent, all this other stuff.

Then there's the stuff that you truly can like truly centralize, which is, okay, yes, investor relations at the end of the day. Yes, like investors want to deal with us as a firm and of course we have people who like, you know, work on the bio fund and other stuff like that, but that makes sense to centralize it because at the end of the day you're calling on the same types of people. Um, you know, core legal, accounting, finance, right, stuff like that. And then I would think of it as like, again, things that are compliance and/or risk-related functions where you don't want one organization to do something that then kind of, you know, makes the whole firm, you know, be brought up in front of some charges for doing something stupid, right? So things of that sort make sense. Now there's always domain, like again, you can do this in different ways, right? You could have it centrally and you could have it dotted line reporting into the different organizations, but at the end of the day, like I would think about it as like IR, compliance, risk type things and accounting, you can centralize those. Everything else over time really becomes domain-specific.

Yes.

Well, look, all these things are hard to manage and this is again, I think the reason why Mark and Ben have made it successful was a couple things. Number one was, um, look, you're absolutely right, which is, you know, you need money to do all these things. And look, you know, uh, we've had the benefit over time at A16Z of being able to have enough of a management fee stream where you can basically do, you know, you can afford to do these things. Now, you also have to get general partners who buy into this, right? Because at the end of the day, and again, you know, I always say this, like the general partners, like no one should feel sorry for the GPs at A16Z, right? Their kids all, you know, have shoes to wear to school and, you know, nobody's like missing meals. But at the end of the day, like if you just took the management fee money that Andreessen Horowitz has, you could pay it all out to the general partners and obviously they would, you know, be even better off than they are today, right? But so part of it is like you have to have people who kind of select into the idea that the value of the firm is bigger than the individual value of what I might derive from compensation on the salary side and hopefully I make it up because the carry that I earn is that much more valuable and the leverage on my time and all those other stuff. So I think that's part of it.

Um, the other thing I think that we did a really good job on on the talent side is you can't make it transactional and you have to basically be there. You have to basically be a partner to the talent, right? And I know that sounds kind of weird, but like it's not good if like we meet some great CFO and then we're like, 'Oh man, this person's great. We got to put him into this really crappy company that's in trouble and like, you know, not tell the truth and stuff like that.' Like the whole idea was, and Jeff Stamps, the guy who set this group up for us, was we got to have a relationship with the talent and we want to basically help them think about how can we help them advance their career. And then if you do that, you establish that trust, then yes, maybe there is a troubled company where you're like, 'Hey, like this is a troubled company. We need your help here, but like hopefully help us fix this and then we're going to put you in.'

Exactly. [laughter] Right. And then we'll introduce you to Ali at Databricks, right? Or something like that. Right. So, but you can't do that unless you kind of have built it that way. I think again where these things go wrong sometimes is they become too transactional and you lose the value of the relationship over time. And again, I think the team has done a really good job of perfecting that.

Yes.

Well, that is a, and that is if you've ever been to our offices or you've read our stuff, you will see that is a centralized cultural tenet of the firm, which is we take a long-term view of relationships. We literally say it that way.

In case my team doesn't show up, in case nobody else shows up, like the team is ready to go, right? And he gets in there, he's like, 'Cameras, lights are already here. Like, why did we bring so much stuff?'

It would have been nice to know that, but it would have been nice to know that in advance. We only do first-class business.

They're related. The first-class business piece, as you probably know, comes from the original J.P. Morgan, right? So that was actually the moniker of the original, like J.P. Morgan was like, 'We only do first-class business in a first-class way.' Um...

Yeah. So that's where that comes from. And, uh, you know, and again, that was something that Mark and Ben adopted. The long-term relationships piece is, I mean, I think you could argue, right, that's an offshoot of that, right? Which is that's part of doing first-class business. But that is also a separate foundational, like, cultural value for the firm, which is, um, and I think you're right, which is, look, it's weird to even say that in the venture world because of course venture is a long-term relationship business. I mean, obviously, look, we're with these companies for 10, 15, sometimes 20 years, right? Who knows, right? I mean, SpaceX, what, just went public a couple weeks ago? I'm pretty sure it was a private company for almost 20 years, 25 years.

So it's not like these are not short-term relationships, but what it really meant, obviously, and again, I think this is where hopefully you see it, is it means you can't be transactional in any of these things and it's got to be true with entrepreneurs. It's also true with LPs. Look, I think, you know, I had the pleasure of like doing, you know, a ton of our LP stuff for, you know, a huge amount of time before we brought in Jen Ka, who now runs it like amazingly well. Um, but I think if you talk to our LPs too, like they're like, 'Wow, you guys do take a very long-term view of this stuff.' And, look, you know, our view, uh, which, you know, I know Ben and Mark echo all the time, is like you have to build long-term. Like, you know, the only way to make sense of these things is you got to recognize that you're building goodwill over time. And look, there are times where you're going to have to draw on that goodwill because you have a bad fund or you have a bad economic cycle or something like that. But if you take a long-term view of that, I think you find that, you know, you have a lot of forgiveness when things like that come up that maybe everyone, you know, isn't excited about.

David was there before, uh, and then he brought on Catherine and then...

Yeah. So the answer is yes. It was like, it was organic was the basic answer, was yeah, like there was no top-down meeting where like, okay, like what should we do next, right? Uh, but it was, and again, this is where again I give Ben and Mark an enormous amount of credit is they give the GPs the freedom to go figure that stuff out, right? So again, they...

It is very unique, right? And this is again, like, look, there's some things like from a governance perspective that are different in the firm, which is again, you've probably heard this, but like, you know, at the end of the day, look, Mark and Ben own the management company, which de facto means at the end of the day, look, you...

Yeah. Like we all, you know, we all serve at their pleasure. It's just, you know, simple way to describe it. Um, and but I think also, but again, and we've been transparent on that. Like so no GP comes in like thinking that they basically aren't in that situation. Obviously how you handle economics is totally different and they've been very, very generous in terms of setting up the economics in a different way. But what I think that governance structure enables us to do these kinds of things, right? Because then what happens is Chris Dixon comes to us and says, hey, like there's this great opportunity in crypto and Ben's like, great, like if you think it's there and you're the subject matter expert and you're going to go like put your entire career on the line for this thing, like yeah, let's go do it. Now again, you know, obviously Mark and Ben ask a lot of questions because they're, you know, smart guys. And so if someone came and said, you know, there's a great opportunity for us to do like Beanie Babies or something, my guess is probably they'd throw them out, you know, throw them out on the floor. But, um, but it's a really, really unique structure in that way and these things happen organically. Yes. American Dynamism. And by the way, some things don't happen. So I'll just give you an example. Like we've had a bunch of people who are like, 'We should be in the private equity buyout world, right?' And this comes up, by the way, every other year.

Is that right? Okay. Yeah. And, uh, and they're like, 'We can make a ton of money.' And Mark and Ben are like, 'Yeah, we can make a ton of money probably.' But like it's just not, that's not brand consistent. That's not who we are. We are technological optimists. Like we believe in the power of entrepreneurship, the power of like the CEO running the company. We can't have a business where we are like simultaneously saying that on one end, on the other end we are like, you know, laying off 30% of the people because we think like we have to make a better EBITDA, you know, number for next quarter or something, right? I'm, you know, being a little bit facetious in terms of that one. But, uh, so there's stuff like that. Like we looked at credit for a long time, um, and maybe that's something that we will do at some point in time, but it wasn't right at the time. Um, I spent a bunch of time, Mark asked me to look at, okay, should we have a public investing fund, uh, like in the, uh, you know, uh, like 2019, right before COVID, we spent a bunch of time. In fact, we were interviewing people for that. And so it's, I say that not to say that like they have not like, you know, basically said like it's carte blanche, do whatever you want, like everything has to fit into the overall strategy. But this idea that people can come up organically, like David Ullrich did with American Dynamism or obviously, you know, what Andrew did with gaming or what's happened on the bio side, like that has all been like driven organically by partners who are like, yes, like this is the right thing. And then, you know, convince Mark and Ben the opportunities there. And then the theory is then great. Okay, like let's build it up over time, right? So just because it's a great idea, like as you've seen like with speedrun, okay, like we can't go, let's not go raise a billion-dollar fund the first time, even though, you know, arguably maybe you could do that if you, you know, just given the LP relationships, let's like prove it out, let's make it work and then, you know, like grow organically, uh, you know, in time.

Yeah, look, I mean, obviously I don't, look, I know Josh well and I have an enormous respect for him and GC obviously is an incredibly great firm. So look, I don't pretend to know exactly what their motivation were in those things. Um, but look, I can understand how they got to those decisions because, you know, uh, you know, if you look at like Josh, he's like, 'Look, we can, we have the relationships. We understand these things.' And, oh, by the way, like, you know, there are natural synergies, I think, between. So like I certainly don't like think it's a crazy idea.

Um, I think a lot of these things come down to, you know, maybe just idiosyncrasies of like what do you want to, what do you want, where do you want to spend your time? And again, at least for right now, and Mark and Ben, uh, Mark has always taught me like consistency is never an objective. Getting it right is the objective. So look, he may wake up tomorrow and say, okay, yes, I've said we'll never do buyout, but now we're doing buyout. So like he reserves the right to do that. He is the leader of the firm.

But I just think it's more like if you ask them like what do you want to do and where do you want to spend your time every day? I think their view is, look, I'd much rather spend my time with some really interesting entrepreneur who's doing something really cool, uh, and like has, you know, whatever a 1% chance of succeeding, but if it succeeds, it changes the world. Like I just think that's part of like who they are as people.

Um, look, that may change. Um, and look, if you had a view that you wanted to be a publicly traded venture capital firm, for example, you would do those other things because at the end of the day, as you probably know, like the public market values like management fee streams more than it does like carry fee streams, right? And so certainty of fee stream becomes more important. So, and I'm not, again, I don't know whether those are aspirations that those individuals have, but I would say like I can tell you that is not an aspiration that Mark and Ben have, having managed a public company in the dot-com era. [laughter]

Yeah. They're like, 'Look, like the last thing we ever want to do is run another public company.'

Exactly. Right. Netscape was a real interesting challenge. Of course, we went through it with LoudCloud and Opsware. And so look, again, now maybe, you know, again, you should...

You should hear from Ben and Mark on this, not me, because, uh, they may say everything Scott said is totally BS and like, you know, don't believe anything he says. But like, you know, at least based on what I understand of them, that's not an objective for how they think about the business.

Yeah.

Yes.

Yes.

Many of us did not think about working in DC. [laughter]

Yeah. So, it's really a couple things. I mean, first of all, uh, you know, the precursor to it was just having been incredibly lucky in my life and being in a situation where like I had the, you know, kind of financial flexibility and, you know, the blessing of Mark and Ben to say, 'Okay, look, like if there's something else that like you think is your highest calling, like by all means, go do that.' Um, so truth be told, like I always had this, it was always on my bucket list to do something in DC. I was literally a public policy major at Stanford. In fact, actually Kevin Warsh is speaking right now probably as we're recording doing his first, uh, press conference as the Fed governor. Uh, we were in the same major together at Stanford. So we've known each other for a very long time. And I spent a summer, uh, you know, as an intern, you know, literally like getting coffee for people and, you know, probably making photocopies at the time, um, in DC. So, but I always thought I would do, I always had an interest generally in policy stuff and I kind of put that on the back burner for literally 35 years.

Yeah. So there's in theory some through line here even though it's, uh, you know, this might just be post-revisionist history, but [laughter] uh, but there is some...

Exactly right. Um, but a lot of it was, look, as you know, as a firm, um, we became more involved in policy, particularly starting with when Chris Dixon started the crypto fund, right? Um, because we recognized obviously the intersection between policy and regulation and all this other stuff. And when we first started the fund, we didn't have like this great team that we have today in DC. So I was the person who like, I raised my hand, I'm like, 'Okay, sure. I'll go testify in front of the Senate or go to the SEC and argue why crypto is, you know, should not be a security and all this other stuff.' And so I kind of scratched my itch a little bit that way. I was the president of the, uh, or excuse me, the chairman of the National Venture Capital Association for a year, which is kind of the trade group for the organization. So I kind of had some at least inkling of this stuff. Uh, later on, of course, like we actually brought real people in who knew what they were doing. So we now have, I don't know, a 20-plus person team in DC who actually understand how to, you know, kind of do these things.

So that was kind of there. And then what happened was, look, as we got involved in, you know, more in politics, and particularly when Mark and Ben decided ultimately that they would, you know, give money to and endorse Trump, uh, you know, in the last election, uh, there was a general call. And so Mark was involved, you know, just through that relationship, there was a general call to people like Mark and obviously David Sacks and others, which is, hey, look, if there are people in the tech industry who are interested in serving the country, like, you know, the president was at least open to, you know, having those types of individuals in addition to obviously the more traditional political players, of course, who, you know, often been in. So that's how I got involved. Uh, so Mark called me up and so I sent an email. The very true story is I sent an email to Mark in like June of '24, actually to Mark and Ben, and I was like, hey, if this works out and you guys are comfortable with it, like I might want to put my hat in, you know, the ring to see if this works out. And so they're like, yeah, okay, like whatever. And I kind of forgot about that email.

Uh, and then Mark comes back to me in September. He's like, okay, I assume you were serious. I've just introduced you to a guy named Howard Lutnick, who of course now is the Commerce Secretary. Howard at the time was running the transition team. Uh, and he's like, you need to go to New York and see him if you like are serious about this. So I got on a plane and went to see Howard. And if you've ever met Howard, you know him. He's like, you know, one of the most gregarious guys. And he's like, yeah, this is great. He had this like, you know, I don't even know how big it was. He had this like massive org chart on his desk.

Super charismatic. And he's like, you know, this is what you should do. And at the time it actually was not OPM. So, uh, he wanted me to go, uh, he's like, you should go, uh, work for a guy named Russ Vought, uh, who's now the head of OMB, the Office of Management and Budget. He's like, that would be a perfect fit for you. And so anyways, that's how I got into the process.

You know, the process is a very long and torturous process, uh, where you end up, uh, actually not in Mar-a-Lago but in a like terrible office building in West Palm Beach or I don't even know where it is, but somewhere like that. Where, um, and I'll never forget because the first time I went out there, I went, it was last minute. They're like, oh, by the way, we're, you know, interviewing for people. Like, can you be here tomorrow? And, uh, and so I tried to get a hotel and I ended up getting one of these, um, I don't even know what you call them, like one of these like Western family hotels or something where it turns out it's like, you know, every room is in theory a suite, but...

Uh, so I was the only one who was dumb enough to stay at this hotel, but I stay at the hotel and it's literally like a combination of like a bunch of kids running around having like, you know, in the pool. And then at the same time, and then I got up the next morning, I remember I got up early to go to the gym and there was like 50, like 75-year-old women doing water yoga or something in the pool. And [laughter] I was like, oh my god, like what have I gotten myself into? And it was nowhere near like Mar-a-Lago or anything like that.

So anyways, yeah, so there were a bunch of people. The nice thing about it was once I got there, there were a lot of people in the tech community who were actually part of the interview process. So I got to, you know, I actually interviewed with, there was kind of, you know, there was a bunch of obviously more Republican political people, but then there was a bunch of people in the general, you know, Elon orbit, right, who were interviewing for that. So it was at least more comforting once I finally showed up like, okay, I actually know some of these people and this is not completely crazy.

Yeah. But it was fun. My favorite story, that was my very first interview was with actually, uh, Peter Navarro, uh, who is, um, uh, now he's a, I don't know what his official title is, but he's kind of, you know, been a longtime, uh, Republican and he had just like, I think he had gone to jail or something for a period of time, uh, for some Trump 45-related, uh, thing. Um, and, uh, with, uh, Katie Miller, who's Stephen Miller's wife, who again, um, uh, also was part of the administration. And, uh, like the first thing that Peter says to me is he's like, 'Oh, like you're another one of these Silicon Valley guys.' He's like, 'How many Chinese nationals work at your company?' And I said, 'Look, I don't know the answer, but you know, I'm sure it's like greater than zero.'

And he's a big, you know, tariff guy as you know and everything. And he's like, you know, I can't, I don't understand how you do that. Like, how can you guys possibly do this? This is crazy. Like, you have like, you know, spies in your companies and stuff. And I was like, look, this is not going well.