How Venture Capital Works: Term Sheets, Unicorns, and Why Most Startups Fail

In 1994, Amazon needed $1 million to survive. Jeff Bezos had used his parents' savings and was burning cash fast. Kleiner Perkins, a prominent Silicon Valley venture capital firm, passed. Nick Hanauer, a Seattle businessman, invested $40,000 for approximately 2% of the company. That stake, held through Amazon's IPO in 1997, was worth roughly $30 million — a 750x return. Amazon's total return from that investment to today would be in the billions.

Educational only — not financial advice. Investing involves risk of loss; past performance does not guarantee future results. Consult a licensed financial advisor before making investment decisions.

Venture capital is a business model built on stories like this: identify companies before anyone else can price them accurately, invest early, and wait for the handful that will generate returns so enormous they cover all the failures. Understanding how this model works — and its distorting effects on startup culture, valuations, and the technology industry — requires digging into both the math and the incentives.

Key takeaways

• VC is a power law business: a handful of investments in any portfolio generate essentially all the returns; the rest typically fail or return little.
• Fund structure drives behavior: VCs manage other people's money (limited partners) and earn carried interest (20% of profits) which creates specific incentives.
• Term sheets are more complex than valuations suggest: preferred stock, liquidation preferences, and anti-dilution provisions determine what founders actually receive.
• Equity dilution is cumulative: each funding round dilutes earlier shareholders; founders who raise too much can end up owning a small fraction of their company at exit.
• Unicorn valuations can be misleading: a "$1 billion valuation" means very different things depending on the preferred stock terms attached to that valuation.

The Venture Capital Fund Structure

Venture capital firms don't invest their own money (mostly). They raise funds from "limited partners" (LPs) — pension funds, university endowments, sovereign wealth funds, and wealthy individuals — who commit capital for typically 10 years.

The VC firm becomes the "general partner" (GP), managing the fund in exchange for two fees:

The management fee: typically 2% of committed capital annually, paid to cover the firm's operating costs (salaries, offices, legal fees). On a $500 million fund, this generates $10 million per year — whether the investments perform or not.

Carried interest: 20% of profits above the initial investment (the "hurdle"). This is where real VC wealth comes from. If a $500 million fund returns $2 billion, the $1.5 billion profit generates $300 million in carried interest for the VC partners.

This structure creates specific incentives. VCs are not particularly motivated by small wins — a portfolio company that returns 2x on a $10 million investment generates only $2 million in carry. A company that returns 100x generates $200 million. This pushes VCs toward high-risk, high-reward investments and away from steady, profitable small businesses.

The Power Law of Returns

The fundamental mathematical reality of VC is the power law distribution of returns. Research by venture capitalist Peter Thiel, later formalized in studies by NBER and others, consistently shows that in any VC fund:

• 30-50% of portfolio companies go to zero
• Another 30-40% return less than the invested capital
• 10-20% return modest multiples (2-5x)
• 1-2 companies generate most or all of the fund's profits

This is not a failure of the model — it's a prediction of it. Venture-backable startups are, by definition, trying things that haven't been done before. Most new things fail. The ones that work often work enormously. A VC who makes 20 investments should expect 15-18 of them to significantly underperform and is planning accordingly.

The implication for founders: VCs need home runs, not singles. A startup that grows steadily to $10 million in revenue and stays there may be a perfectly good business, but it returns almost nothing in VC terms. This is why VCs push founders toward "go big or go home" strategies — it's not irrational; it reflects the actual math of how VC funds generate returns.

How a Startup Funding Rounds Works

A typical venture-backed startup goes through several rounds:

Pre-seed/Seed: $500,000 to $5 million, often from angel investors (wealthy individuals) or small VC funds. At this stage, the company typically has an idea, maybe a prototype, and a founding team. Valuations are speculative.

Series A: $5 million to $20 million, typically when the company has demonstrated early product-market fit. A proper VC firm leads, negotiates terms, and takes a board seat.

Series B: $20 million to $100 million, scaling a proven model. Growth metrics (user growth, revenue growth) are paramount.

Series C and beyond: continued scaling. At this point, companies may have hundreds of millions in revenue. Late-stage investors overlap with public market investors; terms start resembling those of public offerings.

Each round involves issuing new shares, which dilutes all existing shareholders proportionally. A founder who owns 80% after the seed round might own 45% after Series A, 25% after Series B, and 12% at IPO. The absolute value of their shares may have grown enormously (12% of a $5 billion company is $600 million), but the percentage continues declining.

What a Term Sheet Actually Contains

A term sheet is a non-binding document that outlines the major terms of an investment. Founders often focus on the headline valuation number ("pre-money valuation") and miss provisions that can significantly affect their actual payoff.

Preferred stock: VCs almost always receive preferred stock, not common stock. Preferred stock has additional rights that common stock (held by founders and employees) lacks.

Liquidation preference: the most important term. A "1x non-participating" preference means VCs get their money back first in a sale, before common stockholders receive anything. If a company raises $20 million and sells for $25 million, VCs take $20 million and founders split $5 million. A "participating preferred" is worse: VCs get their money back AND share in remaining proceeds alongside common stockholders.

Anti-dilution protection: if a later round values the company lower than the current round ("down round"), anti-dilution provisions adjust the conversion ratio of VC shares upward, further diluting founders and employees.

Pro-rata rights: allow investors to participate in future rounds to maintain their ownership percentage.

Board representation: most Series A investors require at least one board seat, giving them formal oversight of the company.

The headline "valuation" of a startup tells you much less than these terms. Two companies both raising at "$100 million valuation" can have very different actual values for founders depending on liquidation preferences and other terms.

Unicorn Economics and the "Valuation" Problem

A "unicorn" is a venture-backed startup with a valuation exceeding $1 billion. The term, coined by VC Aileen Lee in 2013, was meant to signal rarity. By the early 2020s, there were over 1,000 unicorns globally.

The problem with unicorn valuations is how they're constructed. A startup's "valuation" is set by the price paid for preferred shares in a financing round. If an investor pays $100 million for 10% of a company's preferred shares, the company is "valued at $1 billion." But that preferred stock often comes with significant protections (liquidation preferences, ratchets) that make it worth more than common stock on a per-share basis.

Professor Ilya Strebulaev at Stanford Business School studied 135 unicorn companies and found that, accounting for preferred stock rights, the average unicorn was overvalued by 48% on a fully diluted, identical-security basis. A company "valued at $2 billion" might be worth about $1 billion if all shares were equal.

This matters at exit. Many unicorn-era companies that went public via SPAC or IPO during 2020-2021 saw their public market values collapse 70-90% from peak private valuations. The correction reflected markets applying public-company valuation methodology to companies that had been valued using venture-friendly metrics.

Why Most Startups Fail

The commonly cited statistic is that 90% of startups fail. The reality is more nuanced: failure rates depend heavily on how you define failure and what stage you're measuring.

Among venture-backed startups, studies suggest:

• About 25-30% fail completely (no return of capital)
• About 25% are acquired for modest amounts (below the invested capital)
• About 30-35% return capital with modest gains
• About 10% produce substantial returns

Common causes:

• No market need: the #1 failure reason in post-mortems. Founders build solutions to problems that aren't painful enough for customers to pay to solve.
• Cash timing: good companies die because they run out of money during fundraising gaps
• Team issues: co-founder breakups, inability to hire key talent, or leadership failure as the company scales
• Competition: being wrong about competitive dynamics, especially when a large well-funded competitor enters
• Product-market fit never achieved: iterating too slowly or not enough to find the right product for the right customers

The venture model tolerates and even expects these failures. From a societal perspective, the question is whether the VC model allocates capital efficiently or whether it creates incentives for "fake it till you make it" cultures, burn-rate excess, and valuations untethered from economic reality. The answer, honest observers acknowledge, is some of both.

FAQ

Q: What's the difference between venture capital and private equity? A: Venture capital invests in early-stage, unproven companies, typically taking minority stakes and betting on rapid growth. Private equity typically acquires majority or full control of established, profitable businesses, often using debt leverage, with the goal of operational improvement before resale. Private equity targets predictable returns; venture capital targets extreme returns from rare successes.

Q: Do VCs actually add value beyond money? A: Research is mixed. Some studies find that top-tier VC involvement meaningfully helps companies — through networks, operational advice, and credibility. Other studies find minimal value-add beyond the capital itself, with VC "help" sometimes actively harmful through growth pressure. The honest answer is that it varies enormously by firm and partner.

Q: What is "equity dilution" and should founders worry about it? A: Dilution means your percentage ownership decreases as new shares are issued. It's not inherently bad — owning 10% of a $1 billion company is better than owning 80% of a $1 million company. The concern is excessive dilution through too many rounds at insufficient valuations, or dilutive terms (like anti-dilution provisions and participating preferred) that shift value away from founders at exit.

Q: What's the difference between valuation and value? A: Valuation is a negotiated price at which shares change hands in a specific financing round. Value is what you'd actually receive in a sale or at IPO. For early-stage startups, the two diverge enormously. For late-stage or public companies, they converge, though they're never identical.

Q: Can you get VC funding without being in Silicon Valley? A: Increasingly yes. While Silicon Valley still accounts for a disproportionate share of U.S. VC investment, major ecosystems exist in New York, Boston, Austin, London, Berlin, and Beijing. Remote work and Zoom fundraising normalized during COVID have made geography less determinative, though relationships and networks still matter.

---

Read next:

• Countries with the Most Venture Capital Investment (2025 Ranked)
• How the Stock Market Works: A Complete Guide
• Compound Interest Explained: The Power of Exponential Growth