20VC: Is Seed Investing Dead Without a $1BN Fund? | Does Ownership and Price Matter When Companies Can Be $1TRN Exits | Are AI Revenue Numbers Real and What to Watch Out For with Venky Ganesan, Menlo Ventures

What It Covers

Venky Ganesan of Menlo Ventures joins Harry Stebbings to examine how AI is reshaping venture fundamentals: seed round sizes hitting $100M+, ownership stakes compressing below 2%, AI revenue metrics being gamed, and why IRR now matters more than cash-on-cash returns in an era where the Magnificent Seven tax every startup's P&L.

Key Questions Answered

• •Seed investing as options strategy: Treat every seed check as an option bet to identify outliers, not a conviction investment. Maintain enough positions to statistically guarantee one outlier, then aggressively concentrate capital only when quantitative revenue data confirms a company is breaking out. Menlo expects 60% dilution from first check to exit, so initial ownership must account for that compression from the start.
• •AI revenue metrics red flags: When investors weight any single metric heavily, founders engineer it. Watch for contracted annual revenue that isn't live, run-rates extrapolated from single best-day performance, and manipulated net revenue retention achieved by deliberately starting with small purchase orders then upselling. The tell is whether founders optimize for terminal business value versus valuation markups at the next financing round.
• •IRR over DPI in the AI era: Venture must now beat public market alternatives by 1,000+ basis points to justify illiquidity. Every AI startup pays a tax to Nvidia, a hyperscaler, and potentially a foundation model provider — all investable at zero fee via index funds. This structural cost burden means slow-growth SaaS-style businesses, even with high revenue quality, cannot generate the velocity required for competitive venture returns.
• •Position sizing ladder strategy: Rather than deploying large ownership stakes early, establish a small seed position to gain information rights and board access, then ladder up capital as outlier status becomes data-confirmed. Menlo reached 20% of fund concentration in Anthropic by building incrementally, not via a single large early check — a critical distinction between informed concentration and speculative overexposure.
• •Downside protection rationalization is dangerous: Assuming strategic acquirers will rescue investments at $1-3B floors mirrors .com-era thinking, when Nortel and Lucent paid billions for revenue-free teams using inflated stock. Acquirers target founders, not cap tables, and can hire founding teams directly without paying investor liquidation preferences. Structured acquisition deals frequently disadvantage passive investors while protecting founders and strategic buyers.
• •LP allocation signal: Performance metrics are a five-to-seven year lagging indicator and reveal nothing about future returns. Instead, LPs should contact founders of successful AI companies and ask which partners they respect and would take capital from again even if not needed. Firms without multiple partners appearing on those unprompted founder recommendation lists lack the forward-looking signal that past IRR numbers cannot provide.