← All episodes

Gokul Rajaram: How to Analyse for Durability and Defensibility in a World of AI

| 6 products mentioned
Watch on YouTube venture capital software defensibility ai disruption competitive moats product strategy business model optimization market sizing

Gokul Rajaram, an operator-turned-investor who previously led product at Google, Facebook, Square, and DoorDash, discusses how to identify durable and defensible software companies in an era of AI disruption. He introduces the "eight modes" framework—data, workflow, regulatory, distribution, ecosystem, network, physical infrastructure, and scale—arguing that companies scoring four or more across these dimensions are well-positioned to survive commoditization pressures that have decimated public software valuations.

Key takeaways
  • Companies need multiple interconnected products to achieve billion-dollar valuations, as each product should naturally emanate from the core offering and serve different purposes (profit vs. retention).
  • The "eight modes framework" assesses software durability: data, workflow, regulatory, distribution, ecosystem, network, physical infrastructure, and scale—companies with four or more modes are defensible.
  • Retention metrics (gross retention and net revenue retention) are more important indicators of business quality than raw growth rates, as many AI-powered products experience high churn from "tire kickers."
  • Vertical software companies must own the full stack to reach $10B+ valuations, not just one function within an industry, and should target the total addressable spend including both software budgets and human labor replacement.
  • Brand has weakened as a competitive mode for B2B companies due to increasing data portability and switching cost reduction, forcing companies to rely instead on deeper workflow embedding and proprietary data assets.
  • For Series A investors, concentrate on fewer companies with double-digit ownership targets rather than deploying across many positions; reserve 35% of capital for follow-on rounds to double down on winners.

Recommendations (3)

First Round Capital
First Round Capital recommends

"first round capital and benchmark where I win I win an LP in both and then green oaks first round would be the seed"

Gokul Rajaram · ▶ 1:13:03

Benchmark
Benchmark recommends

"first round capital and benchmark where I win I win an LP in both and then green oaks"

Gokul Rajaram · ▶ 1:13:03

Green Oaks
Green Oaks recommends

"green oaks first round would be the seed bench with the sea and green oaks would be the would be the growth"

Gokul Rajaram · ▶ 1:13:07

Mentioned (3)

OpenAI
OpenAI "you have of course Daario and and and and the open AI folks but then you have a bunch of other la..." ▶ 1:08:30
Figma
Figma "Figma. Uh five of between 500 and 1,000x at the time of IPO, but it has sadly gone down since then." ▶ 1:15:24
Thiel Fellowship
Thiel Fellowship "Obviously, he has the Thiel Fellowship and a preference for young ambitious founders" ▶ 1:16:35