Zoom went public at a $16B valuation in 2019. By 2021: $160B. Today: back to $18B. The companies that survive the hype cycle are often the ones worth owning long-term.
Most people think VCs invest in 30 companies hoping 3-4 succeed. The real math is way more brutal.
I analyzed the portfolio construction of a typical $100M early-stage fund. Here's what actually needs to happen:
The fund invests $2-4M in 25 companies over 3 years.
For a 3x return to LPs:
β Fund needs to return $300M total
β Minus $20M in management fees over 10 years
β Minus 20% carry to GPs
β Net target: $320M from portfolio
The reality check:
β’ 15-18 companies will return 0-1x (total loss or breakeven)
β’ 4-6 companies will return 2-5x (decent but not fund-moving)
β’ 2-3 companies must return 20x+ to carry the entire fund
That means if you invest $3M in a seed round, you need 2-3 companies to exit for $60M+ just to hit your return targets.
This is why VCs obsess over "fund returners" and pass on companies that "only" have 10x potential. It's not greed β it's math.
When a VC says your $50M exit potential isn't big enough for their fund, now you know why.
What surprised you most about VC fund math?
I asked 20 founders who raised Series A in 2025 what their biggest mistake was during fundraising.
18 of them said the same thing: "I optimized for valuation instead of finding the right investor."
Founders who've been through it β what's the one thing you'd do differently in your next fundraise?
Airbnb had 40,000 nights booked in all of 2009. They're now booking 40,000 nights every 17 minutes. Growth isn't always linear β sometimes it's explosive once you find product-market fit.
Brian Chesky pitched Airbnb to 7 VCs in 2008. All 7 said no.
Their reason? "The market is too small. Who wants to stay in a stranger's house?"
Chesky had $1,000 left in his bank account. Instead of giving up, he did something brilliant:
β Sold custom cereal boxes during the 2008 election ($30k revenue)
β Used that money to keep Airbnb alive for 3 more months
β Built a waitlist of 2 million users with zero marketing budget
β Went back to investors with actual demand data
The same VCs who rejected them in 2008 led their Series A in 2009.
What changed wasn't the idea β it was the proof. Chesky stopped pitching the vision and started showing the reality.
Today Airbnb is worth $75B. Those 7 VCs who passed? They missed a 15,000x return.
The lesson: When investors say your market is too small, don't argue with them. Show them it's not.
What's the most creative way you've seen a founder prove market demand?
Most people don't realize how venture capital fund economics actually work. I broke down a typical $100M fund to show you the math.
Here's what shocked me:
The GP only keeps 20% of investment profits. The other 80% goes to LPs (pension funds, endowments, etc.).
But here's the twist β GPs collect 2% management fees annually for 10 years. That's $20M guaranteed, regardless of performance.
So on a $100M fund:
β $20M goes to management fees (salaries, operations)
β $80M gets invested across 25-30 companies
β The fund needs to return $300M+ for GPs to make serious money
Why $300M? Because carry only kicks in after LPs get their $100M back. Then the GP gets 20% of profits above that.
This explains why VCs are obsessed with 10x+ returns. A $100M fund making 3x ($300M) generates $40M in carry. Split among 3-4 partners over 10 years.
The real money comes from the next fund. Successful GPs raise bigger funds with higher management fees. That $100M becomes $300M, then $500M.
This is why fund size matters so much β and why some VCs struggle when they scale too fast.
What surprised you most about VC economics?
I asked 20 founders who raised their Series A in 2025: "What's one thing you wish someone had told you before your first VC meeting?"
Most common answer: "VCs invest in markets first, teams second, product third β but founders pitch it backwards."
What would your answer be?
Brian Chesky was broke, selling cereal boxes to pay rent, and had been rejected by every major VC in Silicon Valley.
Today, Airbnb is worth $75B. Here's the rejection-to-IPO story that changed how I think about persistence:
**The Setup (2008)**
β 3 roommates in SF couldn't afford rent
β Democratic National Convention was in town, hotels sold out
β They inflated air mattresses, charged $80/night for "air bed & breakfast"
β Made $1,000 in a weekend
**The Rejections (2008-2009)**
Fred Wilson passed: "I couldn't wrap my head around air mattresses in strangers' homes"
7 other top VCs said no. The reasons:
β "Market too small"
β "Who wants to stay with strangers?"
β "Hotels will crush you"
Chesky was so broke he lived on $1 cereal (Obama O's and Cap'n McCain's they made for the election).
**The Persistence**
Paul Graham at Y Combinator gave them $20K and one piece of advice: "It's better to have 100 people who love you than 1 million who sort of like you."
They focused obsessively on hosts in NYC. Chesky personally photographed listings. They built tools hosts actually wanted.
**The Breakthrough**
By 2010, bookings were growing 2x month-over-month. Suddenly, the same VCs who passed were calling.
Greylock led their Series A at a $7.4M valuation.
**The Lesson**
Chesky didn't change his idea to fit investor feedback. He changed the execution until the numbers were undeniable.
Sometimes the best validation isn't a yes from VCs β it's customers paying you when you're selling cereal to survive.
What's the longest you've seen a founder persist before breaking through?
Airbnb was rejected by 7 VCs before raising their Series A. One investor said "the market isn't big enough."
That Series A was $7.2M at a $60M valuation. Airbnb is now worth $80B.
The lesson: market size is often limited by imagination, not reality.
Reddit just went public at a $6.4B valuation. Here's what early investors actually made β and it's a masterclass in patience paying off.
If you invested $100K in Reddit's 2005 seed round at a ~$10M valuation:
β 2005: $100K buys 1% of Reddit
β 2006: Sold to Conde Nast for $10-20M (2-4x return)
β 2011: Conde Nast spins out Reddit as independent company
β 2012-2021: Multiple funding rounds dilute but grow the pie
β 2024: IPO at $6.4B valuation
That original 1% stake? Worth $64M at IPO. A 640x return over 19 years.
But here's the twist: most seed investors never saw those returns. They got bought out in the Conde Nast acquisition after just 1 year.
The real winners were the employees who got equity in the spinout and VCs who bought in during the 2012-2014 rounds when Reddit was "just a message board."
Lesson: Sometimes the best exits are the ones you don't take. But try explaining that to LPs waiting 2 decades for a return.
What's the longest you've held a startup investment before seeing an exit?
I asked 50 founders who raised Series A in 2024-2025: "What's the one thing you wish you knew before your first institutional round?"
The #1 answer: "How much time you'll spend managing investors after the money hits your account."
Most thought fundraising ended when the wire cleared. Reality: investor updates, board prep, and stakeholder management can eat 10+ hours per week.
Founders who've been through it: what would you add to this list?
Tesla was rejected by every major VC firm in Silicon Valley in 2004. They raised $7.5M from individual angels instead.
Sometimes the best investors are the ones who say yes when everyone else says no.
I looked into how Anthropic just raised $4B from Amazon at a $18.4B valuation β and the math reveals something fascinating about AI economics.
Amazon led the round but this wasn't typical venture investing. Here's the real structure:
β $4B investment over multiple years
β Anthropic commits to using Amazon's cloud infrastructure
β Amazon gets exclusive access to Anthropic's models for AWS customers
β Revenue sharing agreement on enterprise deals
The math that matters: Anthropic burned ~$2.7B in 2024 training Claude but generated only $200M in revenue. That's a 13.5x burn-to-revenue ratio.
Compare that to OpenAI: $3.4B burn, $3.4B revenue in 2024. A 1:1 ratio.
Why did Amazon pay 18x revenue for a company burning cash this fast? Because they're not buying a SaaS company β they're buying a moat.
Every dollar Anthropic spends on compute flows back to Amazon Web Services. Every enterprise customer becomes an AWS customer. Amazon essentially created a $4B customer acquisition machine.
The real winner? Anthropic's early employees. Series A shares from 2021 are now worth 92x their original value.
Is this the new playbook for big tech β invest in AI companies that become your biggest customers?
I analyzed Anthropic's $4B deal with Amazon and the math is fascinating.
Amazon invested $4B for a minority stake in the AI company. But here's what makes this deal unique:
Amazon gets:
β Equity position in Anthropic
β Anthropic commits to use AWS as primary cloud provider
β Access to Anthropic's AI models for Amazon's products
β Partnership on custom AI chips
Anthropic gets:
β $4B in funding without giving up control
β Guaranteed cloud credits (essentially Amazon paying itself)
β Access to Amazon's chip development resources
β Strategic partnership with a trillion-dollar company
The clever part: Amazon structured this as both an investment AND a commercial agreement. They're betting $4B that Anthropic becomes the Google of AI, while also ensuring Anthropic spends a chunk of that $4B right back on AWS services.
For context, this values Anthropic at roughly $18-20B based on the investment size and reported stake percentage.
This is the new playbook for Big Tech AI investments: write massive checks, but structure deals that create strategic moats and revenue loops.
What other Amazon-Anthropic partnership benefits am I missing?
I asked 15 VCs what the biggest mistake founders make in Series A pitches is.
13 of them said the same thing: "They pitch the product, not the business."
Translation: They talk about features and roadmaps instead of unit economics, market size, and path to profitability.
What's the biggest pitch mistake you've seen or made?
Brian Chesky got rejected by 7 VCs in a row before Airbnb became worth $75B. But it wasn't just the rejections β it was what he did between them.
2008: Airbnb was making $200/week. Investors called it "a Web 2.0 solution for Web 1.0 problems." Paul Graham at Y Combinator was their only believer.
The breakthrough wasn't a new product feature. It was Chesky personally visiting their biggest market: New York.
β He knocked on hosts' doors unannounced
β Took professional photos of their spaces himself
β Redesigned listings with better copywriting
β Revenue in NYC jumped 2-3x immediately
When Chesky went back to VCs, he had different data. Same company, same market size, same "crazy" idea about staying in strangers' homes.
But now he had proof that small changes drove massive growth. That the founder would do things that didn't scale. That they understood their users intimately.
Greylock led their Series A three months later.
The lesson: VCs don't just invest in markets or products. They invest in founders who obsess over the details that move the metrics.
What's the most unconventional thing you've done to understand your customers?
Most VCs I know track their portfolio companies in Excel spreadsheets and quarterly emails. Wild that a $2B+ industry still runs on manual updates.
That's why we built automated portfolio tracking at Raziel. Your founders update once, you see everything in real-time.
I looked into why Anthropic just raised $4B from Amazon at a $18.4B valuation β and the deal structure is wild.
Most people see the headline: "AI startup raises massive round." But the details reveal something bigger.
The breakdown:
β Amazon invested $4B for ~22% ownership
β But it's not equity β it's convertible notes
β Anthropic committed to use Amazon's cloud infrastructure
β Amazon gets exclusive access to Anthropic's foundation models
This isn't just a funding round. It's a strategic alliance disguised as venture capital.
Compare this to OpenAI's Microsoft deal: $10B investment, similar cloud commitments, but Microsoft got board seats and deeper integration rights.
Anthropic kept more control but gave up potential upside. If they hit a $100B valuation at exit, Amazon's $4B investment becomes worth $22B.
The real winner? Amazon's cloud business just locked in a customer spending hundreds of millions annually while betting on the next frontier of AI.
Is this the new playbook for mega-rounds β strategic investors trading capital for exclusive partnerships?
I asked 20 founders who've raised Series A+ the same question: "What's one thing you wish someone had told you before your first fundraise?"
The most common answer wasn't about pitch decks or valuations.
It was: "Fundraising takes 2-3x longer than you think, and it will consume your life for months."
Founders who've been through it β what would you add to this list?
Netflix turned down a $50M acquisition offer from Blockbuster in 2000. Blockbuster filed for bankruptcy 10 years later. Netflix is worth $240B today.
Sometimes the best deal is the one you don't take.