How a VC Fund Works

In 2004, a venture capitalist named Peter Thiel wrote a $500,000 check to a college dorm room project called TheFacebook. When Facebook went public in 2012, that investment was worth over $1 billion. That one bet, in a fund that probably made dozens of other bets that went nowhere, is the whole point. Not a cautionary tale. A masterclass in the math.

The short answer

A venture capital fund pools money from wealthy institutions and individuals, invests it in early-stage startups over several years, and returns profits to investors roughly a decade later. The fund makes money not on the average investment, but on the one or two that become enormous. Everything else exists to fund the pursuit of those outliers.

The full picture

LPs and GPs: who has the money and who bets it

A VC fund has two kinds of participants.

Limited Partners (LPs) are the investors: university endowments, pension funds, insurance companies, sovereign wealth funds, and very wealthy individuals. Harvard’s endowment, managed by Harvard Management Company, is an LP in dozens of VC funds. So is CalPERS (the California pension fund for public employees), which manages over $500 billion in assets and invests across alternative asset classes including private equity. These institutions write checks ranging from $5 million to $500 million into a fund, hand control to the GPs, and wait.

General Partners (GPs) are the venture capitalists themselves: the people who source deals, negotiate investments, sit on boards, and decide where the money goes. A fund might have 3 to 10 GPs, supported by a team of analysts and associates. Sequoia, a16z, and Benchmark are examples of GP firms.

The relationship is simple: LPs bring the capital, GPs bring the judgment. LPs have limited liability. GPs run the show.

The fund structure: size, fees, and the 10-year clock

A typical VC fund looks like this: a pool of capital (say, $500 million), deployed over 3 to 4 years into 20 to 30 startups, with follow-on investments over the next few years, and final returns distributed to LPs by year 10.

The 10-year structure is real. When a VC invests in your company, they have roughly a decade to turn that bet into a return. After that, the fund winds down. This is why VCs are often pushing portfolio companies toward exits (IPOs or acquisitions) even when founders would prefer to stay private longer.

The fee structure is the famous “2 and 20” — a structure documented extensively in VC literature. GPs charge:

• A management fee of 2% of committed capital per year. On a $500 million fund, that’s $10 million annually. This pays salaries, office costs, legal fees, and travel. It’s not profit; it’s operating costs.
• Carried interest (or “carry”) of 20% of profits above the return threshold. This is where GPs actually get rich.

The management fee keeps the lights on. The carry is the incentive.

How a VC actually makes money: carry

Suppose a $500 million fund invests in 25 companies. Eight fail completely. Twelve return modest amounts. Four do well, returning 5x to 10x. And one hits the jackpot: a $10 million investment becomes worth $500 million, a 50x return.

That one investment returned the entire fund by itself.

The total return might be $1.5 billion, a $1 billion profit. The GPs take 20% of that profit as carry: $200 million, split among the partnership. For a team of 5 GPs, that’s $40 million each. This is why top VCs at top funds are extraordinarily wealthy. It’s not the salary. It’s the carry.

The power law: why one investment needs to return everything

The distribution of startup outcomes is not a bell curve. It’s a power law: a tiny number of outcomes generate almost all the value, while most investments return little or nothing. This is why VC returns are driven by outliers, not averages — the math looks nothing like investing in index funds.

Analysis by Andreessen Horowitz, drawing on aggregated Horsley Bridge fund data spanning 1985 onward, found that roughly 6% of investments generate ~60% of all venture returns. The top 1% of outcomes generate something like 40% of all returns.

The implication is counterintuitive. A VC fund’s job is not to have a good batting average. It’s to be in the game when a company like Google, Facebook, or Airbnb is getting started. Missing those deals, even if you had a great average otherwise, means you underperformed.

This is why VCs write checks into companies that sound crazy. The investments that power-law out often look improbable at the time.

The stages: what pre-seed through Series D actually mean

Pre-seed: The earliest stage. Maybe a founding team and a prototype, or just an idea and impressive credentials. Check sizes: $250,000 to $2 million.

Seed: Some early evidence, a product that works, a few customers, initial revenue. Rounds range from $1 million to $5 million.

Series A: The company has demonstrated something real: product-market fit, a user base, a business model starting to work. Rounds typically $5 million to $20 million.

Series B: Scaling what works. More sales, more engineers, expansion into new markets. Rounds: $20 million to $100 million.

Series C, D, and beyond: Companies that are clearly working and growing fast. Pre-IPO rounds can be $200 million or more.

Each round typically involves selling a piece of the company, usually 15% to 25% equity, in exchange for capital.

Why VCs say no so often

A partner at a top VC firm might see 3,000 deals a year and invest in 5. The rejection rate is around 99.8%.

This isn’t because most pitches are bad. It’s portfolio math. Most companies, even good ones, won’t become billion-dollar companies. A fine business with predictable 20% annual growth, healthy margins, and a happy team might return 3x to 5x on investment, which is a great outcome for the founder and a disappointing one for a VC fund that needed 50x.

VCs say no not because you’ve failed, but because your particular opportunity doesn’t fit the return profile of their fund model.

What happens when a VC fund fails

The fund model has a built-in protection for GPs that most people don’t realize exists. Even if a fund performs poorly, the GPs have been collecting 2% management fees for a decade. On a $500 million fund, that’s $10 million per year, or $100 million in fees over the life of the fund, just for showing up.

This creates an uncomfortable dynamic. A VC who raised a fund, deployed it into investments that didn’t pan out, and returned nothing to LPs still walked away with millions in management fees. The LPs lost their money. The GPs didn’t.

This is why LP selection of GP track record matters so much. LPs are betting not just on the fund’s current strategy, but on the GPs’ ability to actually generate exits. A GP with two failed funds might struggle to raise a third — but might not, if they have strong enough relationships and a compelling narrative.

There’s also the J-curve problem: VC fund returns look terrible for the first several years. This is the opposite of compound interest, which rewards patience with accelerating gains — VC funds look awful until the exits finally materialize. The fund has deployed capital, companies are spending money to grow, and none of the investments have exited yet. On paper, the fund is worth less than what LPs put in. Only after companies start going public or getting acquired does the fund value start rising steeply. Investors who can’t stomach a decade of apparent underperformance before seeing returns aren’t suited to VC.

Why it matters

Understanding how VC funds work explains a lot of seemingly puzzling startup behavior. Why does your favorite startup keep raising money instead of turning profitable? Because their VC investors need a massive exit. Why do startups grow so aggressively even when it seems unsustainable? Because growth unlocks the next round, and the next round is how the company survives.

VC-backed companies are not optimizing for normal business success. They’re optimizing for the specific return profile that makes venture funds work — typically an IPO or acquisition that lets investors sell into the stock market at a massive premium. That’s rational given where the money came from, but worth understanding if you’re building a company, working at one, or just watching from the outside.

Common misconceptions

“VCs get rich from management fees.” Management fees are operating expenses, not profit. GPs get wealthy through carried interest, 20% of fund profits. A fund that performs poorly leaves GPs with decent salaries and no real wealth creation.

“Getting VC funding validates your business.” It validates that a VC thinks your business could be one of the rare power-law outliers. That’s a specific bet on a specific outcome, not a general endorsement of your idea’s viability.

“Series A is ‘early stage.’” It depends on the era. In 2021, many Series A rounds went to companies with $5 million in annual revenue that had already raised two seed rounds. The label has inflated.