Masters in Business
Masters in Business

Scott Kupor Discusses Technology Startups

Bloomberg Opinion columnist Barry Ritholtz interviews Scott Kupor. Kupor is the managing partner at Andreessen Horowitz where he is responsible for all operational aspects of running the firm. He has been with the firm since its inception in 2009 and has overseen its rapid growth, from three employe

Featured Speakers

Bloomberg HostScott Cooper Guest

Topics Discussed

Episode Summary

Executive Summary: The conversation centers on Andreessen Horowitz managing partner Scott Cooper’s career path and his book on venture capital, using it to explain how VC works, why information asymmetry matters, and how the industry changed as startup capital became abundant. Cooper argues VC returns are driven by a few outsized winners, making branding, domain expertise, board access, and team quality more important than broad diversification. He also discusses Y Combinator, late-stage private capital, valuations, debt vs. equity, and major forces like software, crypto, and the rise of “internet money.”

Main Topics: Scott Cooper’s background and path into venture capital (Priority: 5/5): Cooper recounts moving from Stanford law and banking into operations at LoudCloud/Opsware, then into Andreessen Horowitz as employee number one. His career spans legal, financial, and technical perspectives that inform his VC role. What Sand Hill Road and venture capital represent (Priority: 4/5): He explains Sand Hill Road as the geographic and symbolic center of Silicon Valley VC, tightly linked to Stanford and the startup ecosystem, similar to Wall Street or Music Row. How VC pitch, selection, and returns really work (Priority: 5/5): Cooper describes pitch evaluation, the need to understand VC incentives, and why the asset class depends on power-law outcomes where a small number of investments drive most returns. How the startup funding environment changed (Priority: 5/5): He details the fall in startup costs, the rise of seed funds, and Y Combinator’s role in educating founders and reducing the traditional VC information advantage. Capital abundance, unicorns, and late-stage private markets (Priority: 4/5): The discussion covers the shift from scarce to plentiful capital, larger private rounds, longer time-to-IPO, and the influx of public-market capital into private valuations. Valuation, debt, and the risk of overpaying (Priority: 4/5): Cooper stresses that valuation matters because of follow-on rounds and dilution, while debt can become a trap for startups if it forces repayment before the business is ready. Firm strategy: vertical specialization and post-investment support (Priority: 4/5): Andreessen Horowitz’s model is described as vertically organized teams plus large operating support groups for recruiting, sales, and business development, helping scale the firm beyond traditional partnership models.

Key Arguments: Venture capital is a power-law business: a tiny fraction of investments produce most returns, so the winner selection process matters far more than average accuracy. The industry became less VC-controlled because startup costs fell dramatically and founders gained more leverage through seed funds and Y Combinator. Brand and track record create persistence in VC; top firms keep winning because entrepreneurs, employees, and later investors use the brand as a signal of quality. Diversification is less effective in venture capital than in public markets; limited partners should concentrate capital in top-quartile firms rather than spread it broadly. Valuation cannot be ignored even if an exit could be huge, because the next rounds, ownership dilution, and follow-on capital needs determine whether the investment remains attractive. Most successful startups are judged on team quality and execution, not just product or market idea; failures often arise from management and organizational issues. A16z’s model scales by disaggregating the classic partner role into specialized vertical investors and operational experts who help portfolio companies with hiring, sales, and network access. Crypto and ‘internet money’ are framed as attempts to lower payment friction and enable micropayments, especially across borders and in underbanked markets.

Data Points: Fund size at the time referenced: $7 billion - Cooper notes the firm was previously described this way, then says it is now over $10 billion More recent fund size: About $3 billion raised in the latest fundraise - Cooper says the firm has recently raised another large fund Current firm scale: Over $10 billion - He says this is the updated scale of the firm after the new raise Employees at Andreessen Horowitz: 170 - Cooper says the firm grew from 3 people to 170 employees Partners at Andreessen Horowitz: 15 - He states the firm has 15 partners VC pitches reviewed annually: About 2,000 to 2,500 - He estimates the firm sees this many pitches per year Seed-fund market growth: About 500 new firms in 10 years - He cites the expansion of seed-stage investing Seed capital share: 5% to 6% of total venture capital - He says seed is a growing but still small share of VC capital Median revenue for tech IPOs in 1999-2000: $17 million - Used to contrast the dot-com era with today Median revenue for tech IPOs over the last 10 years: About $170 million - Shows how much larger modern IPO candidates are 1999-2000 tech IPO count: About 750 - Cooper compares this with the much lower recent IPO count 2001 tech IPO count: About 2 - He says there were only around two IPOs in 2001 Recent IPO count over last decade: About 30 to 50 per year - He contrasts modern IPO activity with the dot-com bubble Average VC underperformance versus NASDAQ: 160 basis points below NASDAQ - Cited as of 2017, for aggregate 10-year returns Likely return dispersion in VC: 3,500 to 4,000 basis points - He says good versus bad VC performance can differ by this much Typical VC fund life: 10 years - He describes the standard fund cycle Capital call timing: About 70% called in first 3-4 years - Explains how LP commitments are drawn down over a fund’s life Typical company time from founding to IPO: 10 to 12 years - Up from about 6 to 6.5 years historically Public-market concentration of returns: 4% of stocks drive about 90% of returns - He mentions a study showing public markets also follow a power-law pattern 2018 IPO profitability stat: 15 of the largest 25 IPOs were unprofitable - Used to discuss growth versus profit in public listings SoftBank Vision Fund size: $100 billion - Cited as the largest private-tech fund of its kind VC ownership thresholds: Top partners may sit on 10 to 25 board seats - Used to explain the scaling constraint on VC firms

Pivotal Quotes: "we are wrong more often than we're right" — Scott Cooper: He is explaining the VC decision-making model and why outsized winners matter more than hit rate "information asymmetry is a problem" — Scott Cooper: He is describing why entrepreneurs often feel disadvantaged relative to experienced venture capitalists "software is eating the world" — Mark Andreessen (referenced by Scott Cooper): Used to frame the firm’s ongoing thesis that software continues to transform more industries

Implications: For founders, the message is to understand VC incentives, choose capital carefully, and focus on team quality and execution. For investors, top-tier firms, domain expertise, and brand matter more than broad diversification. For the industry, capital abundance favors founders and prolongs private-company growth.

From the Transcript

That venture capitalists care about? What is it that they are incented to do, and therefore, how are they going to evaluate you? And so, I think the first thing to think about is the most important thing to think about in this business is: we are wrong more often than we're right, which I know is a terrible thing to say. And if your kids came home and got 50% on their test, you'd be pretty unhappy. If you're doing that in venture capital, you're still in the game at least. And as long as you bring up that batting average, why does the hit-miss ratio not matter as much? Much in venture capital as it does elsewhere. Because if we do it right, then the reason it doesn't matter is because 10 to 20% of the companies, hopefully, if we do it right, you're going to earn 25, 50, 100 times your money. And so basically, think of it as the significant winners will basically make up for kind of that 50% of things where you won't get your money back, and then probably the 20%, 30% where you get a little bit of money back, but not enough to make the math work. And so if you think about that, then what that means is if I'm going to invest in a company today, I've got to at least believe.

Scott Cooper · at 11:24

The implication? No, I think the implication is there's just not a level playing field from an information perspective, right? Information asymmetry is a problem. That's right. So, look, we'll do, you know, we've been doing this for 10 years. You know, we've done thousands of deals. You know, a repeat entrepreneur maybe does this five, six, seven times over the course of their life. And so there's just stuff that we know and we see because we see it on an everyday basis. And so the purpose of the book in my mind was: look, if we could demystify that and hopefully level a playing field, then maybe it helps have a better relationship between entrepreneurs and VCs, and maybe it even helps people who wouldn't. Otherwise, I thought about entrepreneurship coming to the business. Huh, interesting. And I have to ask, for the uninitiated, what is Sandhill Road? I've been there. I understand it as a concept. Yeah. The concept, so the concept at a high level is like, you know, what Wall Street is to financial services or what, you know, Music Row is to Nashville country music, right? It's basically, it literally is a road. It is a street. It is very unexciting, as you've probably seen when you've been there. It's a bunch of kind of, you know, fairly drab two-story.

Scott Cooper · at 9:24

And Dreessen's piece, Software is Eating the World. That was 2011, and that turned out to be a fabulous call. So, really, the question is: is software still eating the world? And when does this get replaced by whatever is going to replace software? Or does that just never end? It just keeps going. Yeah. So, I think software is still eating the world, at least where we sit. We think it will continue to eat the world for, I don't know what the time period is, but I don't know. 147. Exactly right. I don't see an end to it at this point in time. And actually, what's interesting now is it's starting to touch a lot of industries that historically it never got to. So now we're starting to see software eating, a little bit of education, a little bit of healthcare, government services, oil and gas markets, energy markets. So there's these very, very weak markets that for a long time kind of were largely untouched. And I think we're still at the very, very beginning phases of it. So it's been our investment thesis for a long time. I think it's going to be our investment thesis for the foreseeable future. So I'm going to throw a curveball at you.

Mark Andreessen · at 53:41
🔓 Sign Up for Unlimited Episode Search

About Masters in Business

Barry Ritholtz speaks with the people that shape markets, investing and business.

View all episodes from Masters in Business