Gokul Rajaram - Lessons from Investing in 700 Companies - [Invest Like the Best, EP.456]

Jack Dorsey reframed the Product Manager role as "Product Editor." The most valuable skill is not generating new feature ideas, but exercising judgment to cut through the noise, simplify complexity, and edit the product down to the essential few things that truly drive customer outcomes.

Schmidt insisted on presenting company strategy using only images, with no text on slides. This constraint forces leaders to distill complex ideas into visceral, memorable concepts that communicate feeling over facts, believing people remember how something made them feel, not the specific words used.

With well-established design systems, companies are finding AI can generate designs effectively. This is causing a strategic shift in headcount allocation, where teams are choosing to hire an additional engineer over a designer, dramatically altering traditional product team ratios from 1:10 to 1:20 (PM to Engineer).

When AI can generate code and designs endlessly, creating "AI slop," the critical human contribution becomes judgment. The key challenge shifts from *building* to *deciding what to build* and evaluating the output's quality and security. The question is no longer "can we build it?" but "should we build it?"

To filter for a bias for action, DoorDash gave candidates a work project: acquire 1,000 customers with $20. The impossible goal wasn't the point; the test was designed to see what candidates would *do*. Their creative and scrappy attempts revealed far more about their mindset than a traditional interview could.

Great product design removes upfront friction. Instead of complex approval processes, Square approved merchants instantly and managed risk per-transaction. Similarly, Google's Sergey Brin killed AdSense's publisher approval system, opting to review sites only after they hit a certain impression threshold, enabling frictionless scale.

The complex ad tech landscape can be boiled down to three viable business models. A company must either 1) own a first-party surface with coveted users (Google), 2) become the best at delivering a specific, measurable result (Applovin), or 3) be the exclusive demand aggregator for large advertisers (The Trade Desk).

AI's rapid capability growth makes top-down product specs obsolete. Product Managers now work bottoms-up with engineers, prototyping and even checking in code using AI tools. This blurs traditional roles, shifting the PM's focus to defining high-level customer needs and evaluating outcomes rather than prescribing features.

Building a fully self-serve product doesn't just cater to small customers. Companies like Square and Figma found that large, sophisticated users often prefer to sign up and explore advanced features on their own. This creates a powerful bottom-up adoption wedge inside large organizations, bypassing traditional top-down sales.

Incumbent SaaS companies like Salesforce are cutting off API access to prevent AI startups from siphoning value. To build a durable business, new AI companies cannot simply be a "system of action" on top of old platforms; they must aim to become the new system of record, which requires building complex data migration tools from day one.

AI agents can easily siphon off value from SaaS products priced on per-seat utility by automating tasks previously done by humans (e.g., support tickets). In contrast, deeply embedded systems of record (ERP, CRM) are insulated by career-limiting switching costs and the immense challenge of migrating timeless, critical data.

Facebook's transformative ad product was born from Mark Zuckerberg translating a major customer's pain point—Zynga's desire to find more high-value "whale" players—into a technical solution. He proposed letting advertisers upload their own customer lists to find lookalikes, a direct, founder-led innovation that became a core feature.