Venture Capital: The Culture of Tech Money
The Million Dollar Question: A venture capital fund backs 20 startups. How many does it typically expect to lose money on?
A) About 2 B) About 5 C) About 10 D) About 12 or moreRead on for the answer.
Venture capital is the only corner of finance built around being wrong most of the time on purpose. A good fund expects the majority of its bets to fail, and it is fine with that, because the whole model is engineered so that one enormous winner pays for all the losers and then some. Almost everything that looks strange about the culture of tech money — the speed, the swagger, the willingness to hand millions to a twenty-six-year-old with a prototype — makes sense once you understand the math underneath it. This is how venture capital actually works, who plays the game, what it costs, and what most people get wrong about it.
What it is
Venture capital is money invested in young, private, high-risk companies in exchange for ownership. Unlike a bank loan, there is no repayment schedule and no collateral. The investor buys equity and is betting the company becomes worth dramatically more later. Most of the time it does not. Occasionally it becomes the next Google, and that single outcome can be worth more than every other investment in the fund combined.
The money is organized into funds. A venture firm — the general partners, or GPs — raises a pool of capital from outside investors called limited partners, or LPs. The GPs then invest that pool into startups over roughly three to four years, help those companies grow, and try to “exit” their stakes through an acquisition or a public offering, usually within a fund life of about ten years. Whatever profits come back get split between the LPs, who supplied most of the cash, and the GPs, who did the investing.
The money arrives in stages, each with its own risk and price tag. The earliest is “seed,” small checks into little more than a team and an idea. Then come the lettered rounds — Series A, B, C and beyond — each one larger, at a higher valuation, as the company proves it can grow. Later still is “growth” or “late-stage” capital, big sums into companies that are already substantial but not yet public. Early rounds are the riskiest and, when they hit, the most lucrative; a seed check into a future giant can return hundreds of times its money, while a late-stage check into the same company years later returns far less because most of the risk — and the upside — is already gone.
It is a big business and getting bigger. The U.S. venture industry now manages around $1.25 trillion in assets, according to the National Venture Capital Association’s 2025 Yearbook, with roughly $307.8 billion in “dry powder” — committed capital that has been raised but not yet invested. That is a lot of money waiting for the right twenty-six-year-old.
Who uses it
It helps to separate the players, because they sit at very different wealth levels and want very different things.
On one side are the limited partners — the people whose money it actually is. These are university endowments, public and corporate pension funds, sovereign wealth funds, insurance companies, foundations, and the family offices of wealthy households (the kind of allocation we cover in Alternative Assets). A pension fund managing tens of billions might place 5 to 15 percent of its portfolio in venture and private equity, hoping the returns beat what public stocks can deliver.
In the middle are the general partners — the venture capitalists themselves. This is not one wealth band but several. A junior partner at a mid-sized firm is a well-paid professional, comfortably in the $1M–$5M net-worth range over time. The founders of the largest firms are something else entirely. Marc Andreessen, co-founder of Andreessen Horowitz, is worth around $1.9 billion according to Forbes. Peter Thiel, who co-founded Founders Fund, was pegged by Forbes at roughly $23.5 billion in early 2025. The gap between an entry-level associate and a billionaire founder-GP at the same kind of firm is one of the widest in finance.
On the third side are the founders — the entrepreneurs taking the money. They are not VC’s customers so much as its raw material. The relationship is covered from their side in The Founder Lifestyle and Tech Wealth.
The cast has been widening, too. Alongside the big-name partnerships, the last decade produced “solo capitalists” who raise sizable funds under a single name, “rolling funds” that let smaller investors commit on a subscription basis, and scout programs that hand checkbooks to well-connected founders and operators to make tiny early bets on the firm’s behalf. The effect is more people writing checks than ever — but the same power law still governs all of them. More entrants does not change the math; it just spreads it across more names.
Why they use it
Each side is chasing something different, and all of it traces back to one idea.
Limited partners use venture capital because, done well, nothing else in a portfolio can match its upside. A pension or endowment cannot get 20x returns from bonds. Venture offers the chance — not the promise — of returns that move the whole fund. They also get access: getting into a respected venture fund can be harder than getting into a respected university, and the best LPs guard their relationships with top firms for decades.
General partners use it for two reasons: the money and the standing. The real wealth in venture comes from “carried interest,” the GPs’ share of the profits, which we will get to. But there is also status. Backing the company that defines a decade is a kind of cultural power that a high salary alone does not buy.
Underneath both motivations sits the power law. In venture, returns are not spread evenly across a portfolio — they pile up in a tiny number of investments. Roughly 60 percent of venture-backed startups fail, and a small handful of winners produce the bulk of all gains. One analysis by the fund-of-funds VenCap, reported by The VC Factory, looked at 11,350 startups backed between 1986 and 2018 and found that only about 1.1 percent of them returned the entire fund that invested in them — yet 90 percent of the funds that made strong returns had at least one of those “fund returners.” The whole game is finding the rare company that pays for everything else.
How it works
The day-to-day looks less like spreadsheet finance and more like a mix of scouting, sales, and judgment under uncertainty.
It starts with sourcing — seeing as many promising companies as possible. Partners mine their networks, founders introduce other founders, and scouts and analysts comb conferences and demo days. Then comes diligence: meeting the team, pressure-testing the market, checking the technology. Because the best deals move fast, this can be compressed into days, which is why pattern-matching — “this reminds me of an early winner I backed before” — plays such an outsized role, for better and worse.
Many firms still run the decision through a weekly partner meeting, often on a Monday, where the people sponsoring a deal pitch it to the rest of the partnership and absorb the skepticism before any money moves. It is part debate, part group judgment, and part ritual — a room of people who know how rarely they are right arguing about which rare thing to bet on next.
If the firm wants in, it issues a term sheet: how much it will invest, at what valuation, and on what terms. A typical early-stage check buys a meaningful minority stake and often a board seat, giving the investor a say in big decisions and the right to invest more in later rounds — “follow-on” capital to protect the stake as the company raises again.
Then everyone waits, sometimes for a decade, for an exit. Most never come. The portfolio is built knowing this. Picture a fund that makes 20 investments: it might write off a dozen entirely, get its money back or a little more on a few, and depend on one or two breakout companies to carry the whole thing. When it works, it really works. Sequoia Capital’s roughly $60 million stake in WhatsApp turned into more than $3 billion when Facebook bought the company — about a 50x return on a single bet. That one deal mattered more than dozens of quiet failures.
The fortunes of the best-known investors were built on exactly these outlier bets. Michael Moritz put about $12.5 million into Google for Sequoia; he is now worth roughly $7.1 billion, according to Forbes. John Doerr made a similar early bet on Google for Kleiner Perkins, anchoring a multibillion-dollar fortune. And in 2005, Jim Breyer led Accel’s $12.7 million investment in Facebook at a $98 million valuation — a stake that helped lift his fortune to $3.8 billion, per a 2025 Forbes profile. In each case, one company did the heavy lifting.
Exits come in two main shapes. The flashier one is an initial public offering, where the company sells shares to the public and the fund’s stake becomes liquid stock it can eventually sell. The more common one is acquisition, where a larger company simply buys the startup. Both can be huge, but they are not evenly available — IPO windows open and slam shut with the market, and in slow years like the recent stretch, U.S. firms have raised far more capital than they have returned through exits, leaving fund managers holding their winners longer and limited partners waiting longer for cash to come back.
Right now, almost all of that searching points in one direction. Artificial intelligence absorbed about 63 percent of U.S. venture dollars in 2025, a record share, per PitchBook, with AI taking more than half of all global venture funding for the first time ever. Andreessen Horowitz raised more than $15 billion across new funds in January 2026 — over 18 percent of all U.S. venture capital raised in 2025 — pushing its assets under management past $90 billion. When the herd moves, it moves together.
What it costs
For limited partners, the price of admission is the fee structure, summed up as “2 and 20.” The firm charges an annual management fee of about 2 percent of the committed capital, plus 20 percent of the profits as carried interest. On a $100 million fund, the 2 percent fee is roughly $2 million a year to cover salaries and operations, paid whether the investments do well or not. The 20 percent carry only pays out once LPs get their money back and clear any agreed hurdle.
That split explains how venture capitalists actually get rich, and it is not mainly the salary. Compensation varies widely by level and firm. According to the 2025 venture capital salary survey compiled on John Gannon’s blog, associates averaged base pay around $126,000, senior associates around $150,000, and partners often earn cash compensation from several hundred thousand into the low millions — with general partners’ total cash running anywhere from roughly $500,000 to $2 million. Real wealth, though, comes from carry. On a fund that returns several times its money, a partner’s slice of that 20 percent can dwarf every paycheck combined. Carry is why the founders of the top firms reach ten and eleven figures while the salary alone never would.
There is a reason to keep the wealth bands straight here. A first-year associate earning a six-figure salary and a billionaire founder-GP are both “in venture capital,” but they live in completely different financial worlds, and the bridge between them is carried interest on a winning fund — something most people in the industry never fully capture.
Geography quietly adds to the cost, too. The center of gravity is still Sand Hill Road in Menlo Park, California, where proximity to the strongest founders, the largest LPs, and rival firms carries a markup that shows up in salaries and office rents alike. Pay for the same role in Silicon Valley can run 20 to 30 percent higher than at a firm in a smaller market. The cost of being where the deals are is part of the cost of doing the deals.
Hidden costs and tradeoffs
The headline returns hide a lot of friction, most of which lands on the limited partners.
The biggest is illiquidity. When an LP commits to a fund, that money is locked up for ten years or more, with no guarantee of when — or whether — it comes back. Early in a fund’s life, returns often look negative: fees are being paid and young companies have not matured, a pattern investors call the J-curve. You have to be willing to look wrong for years.
There is also a trust cost that rarely gets named: most venture commitments are blind-pool investments. When LPs wire their money, they usually do not know which specific companies it will buy — they are betting on the judgment of the partners, sight unseen, for a decade. That is why access and reputation matter so much on the LP side. You are not buying a portfolio; you are buying a relationship with people you believe will find the next outlier.
For founders, the cost is dilution and control. Every round of venture money sells off another slice of the company and often adds investors to the board, with rights that can shape or override the founder’s decisions. A founder who owns 100 percent at the start can be down to a low double-digit percentage after several rounds — still potentially worth a fortune, but no longer fully in charge. Capital is never free.
For the venture capitalists themselves, the tradeoffs are subtler: the long wait for carry to vest and pay out, the herd behavior that pushes everyone into the same hot sector at the same valuations (AI today, crypto and consumer apps before it), and the reputational stakes of public bets. A wrong call is visible, and a missed call — passing on the company that becomes a giant — can follow a name for a career.
What people get wrong
The first myth is that venture capitalists are geniuses who simply pick winners. The power law says otherwise. Even the best investors are wrong most of the time; their edge is portfolio construction, access to the strongest deals, and the discipline to back winners again and again — not a crystal ball. The model is designed to survive being wrong on the majority of bets.
The second myth is that venture is a reliable way to beat the market. As a whole, it often is not. Returns are extraordinarily concentrated — not just within a fund, but across the industry, where a handful of firms capture a large share of all the gains. Get into a top firm’s fund and you may do spectacularly. Land in a middling one and you may underperform a simple index fund after fees. Average venture returns and top-quartile venture returns are almost different asset classes.
The third myth is that the 2-and-20 structure makes venture capitalists rich on fees alone. The management fee keeps the lights on and pays the staff; it rarely builds a fortune. Generational wealth in this business comes from carry on funds that hit, and only a minority of funds hit big enough to matter. The fee is the floor. The carry is the dream.
A related trap is survivorship bias. The stories that get told are the WhatsApps and the Googles — the bets that paid 50 or 100 times over. The hundreds of failed companies in the same funds, and the funds that never returned much at all, do not get profiles written about them. That makes the whole field look more brilliant and more inevitable than it is. For every famous early check into a future giant, there were many similar checks, written with the same confidence, into companies almost no one remembers.
A fourth, quieter misconception is that all this money is patient and visionary. Plenty of it is, but venture is also intensely fashion-driven. When AI commanded close to two-thirds of U.S. venture dollars in 2025, that was not 1,000 investors independently reaching the same conclusion. It was a herd — and herds, in venture as anywhere, sometimes overpay.
Bottom line
Back to the Million Dollar Question: of 20 startups a fund backs, how many does it expect to lose money on? The answer is D — about 12 or more. With roughly 60 percent of venture-backed companies failing and the bulk of returns coming from one or two breakout winners, losing money on most of the portfolio is not a malfunction. It is the model performing exactly as designed.
That single fact explains the culture of tech money better than any profile of a famous investor. The speed, the bold bets, the tolerance for failure, the rush into whatever is hot, the enormous fortunes that sit beside long stretches of looking wrong — all of it follows from the power law and the economics of carried interest. Venture capital is not the art of being right. It is the discipline of being wrong cheaply, often, and on purpose, while staying in position for the one bet that pays for everything.
Related reading: Tech Wealth: How Founders and Investors Live Differently · Hedge Funds and Private Equity: The Other Engine of Modern Finance Wealth · Paths to Millions: How First-Generation Wealth Is Actually Built · Alternative Assets: Investing Beyond Stocks and Bonds · The Founder Lifestyle: Status, Control, and Extreme Ambition
