Anu Hariharan - Lessons in Growth Investing

Principles & Lessons:

• Business model analysis offers better explanatory power than industry classification. Anu explains that focusing on business models (e.g., marketplaces, advertising, SaaS) yields clearer insights than thinking in terms of sectors, especially as tech crosses industry lines. For example, "advertising as a business model, you monetize in a similar way," whether B2B like LinkedIn or B2C like Facebook. This avoids conflating surface-level industry similarities with fundamentally different monetization structures, and allows more accurate reasoning about cost structure, revenue mechanics, and potential margins.
• Not all markets are winner-take-all—contextual advantages drive differentiated outcomes. Anu challenges the overused assumption of "winner-take-most" markets by showing how local execution and problem selection can create parallel, dominant players. DoorDash succeeded not by dominating existing markets, but by entering overlooked suburban regions and tailoring operational strategies—like driver pay structure and merchant integration. “There’s still room for two or three players,” she notes, when scale and selection advantages can be locally replicated or differentiated, contradicting zero-sum thinking.
• The key to evaluating marketplaces is identifying which side is harder to acquire—at each stage. Effective reasoning about platform dynamics requires focusing on the changing bottlenecks in supply-demand matching. Anu recalls Chris Dixon’s lesson: “at different points of your marketplace evolution, you need to figure out which side is hard,” and invest your energy accordingly. Airbnb initially focused on hosts; later, demand acquisition became the constraint. This reflects a more general epistemic principle: explanations must be dynamic and sensitive to phase changes, rather than static generalizations.
• Clarity of thought about scale and trajectory is a non-negotiable quality in later-stage founders. At the growth stage, investors shift from betting on founder-market fit to judging whether founders understand how large their opportunity really is. Anu cites Brex’s founders, who modeled potential market share and unit economics before leaving Stanford: “That clarity of thought is rare.” At scale, reasoning about counterfactuals and future states becomes more valuable than observing current performance—a shift from description to explanation and foresight.
• The upside case must be actively constructed, not passively extrapolated. Anu argues that many investors overemphasize downside diligence and miss grand slams by failing to model plausible upside scenarios. She recalls Marc Andreessen’s advice: “Your bias is going to be to say no… use the same process to figure out the upside potential.” Stripe’s $20B base case missed the compounding effect of layering multiple successful products. Constructing upside requires active imagination constrained by internal logic—not just reliance on historical comparisons.
• Digital transformation is still early—B2B is an especially underappreciated frontier. While consumer e-commerce garners attention, B2B commerce (e.g., wholesale transactions) is vastly larger and far less digitized. “Less than 8% of B2B wholesale is online,” says Anu, despite being a $16 trillion market. This shows that technological change is unevenly distributed not just geographically but within economies—pointing to unexamined assumptions about saturation and progress. It is not enough to know that “digital transformation is happening”; one must ask where it isn’t, and why.
• The hardest technology bets require decomposing risk into multiple dimensions. Investing in a company like Boom Supersonic is often dismissed as too risky, but Anu breaks it down into four distinct risk types: science, engineering, market/commercial, and funding. Boom’s science risk was low (supersonic flight was known), engineering risk was moderate but manageable, and funding risk could be mitigated through non-venture sources. This decompositional approach mirrors good explanatory reasoning: replacing ambiguous general terms like “risky” with specific, testable components.
• Valuation in growth investing is probabilistic and driven by strategic upside, not just comparables. Anu makes clear that valuation—especially at Series C and beyond—is not a rigid equation but an exercise in scenario-weighted thinking. “You don’t ever turn down a company because this is the max valuation you want to go to… if you have conviction.” Entry price matters, but is secondary to the possibility of massive value creation. The real task is to identify mispriced optionality—not to anchor on historical valuation norms. This requires replacing shallow heuristics with explanatory models of how value could emerge.

Transcript