The secret to better AI prototypes: Why Tinder’s CPO starts with JSON, not design

Ravi Mehta, a product advisor and founder of Outpace, previously Tinder's CPO and a Reforge executive-in-residence, discusses the differences between leading at big companies and startups, his product strategy stack, goal-setting with understanding-focused goals, PM competencies, feedback, leadership style, and how AI is changing his coaching company.

01Key learnings

• Startups win on latency, the speed from idea to validated result, not raw velocity; design tests so you can learn in days rather than quarters.
• Early-stage companies lack the traffic for statistically significant experiments, so favor conviction built from enough data over prolonged analysis paralysis.
• Build an early-stage network of founders, angels, and builders early, since large-company connections often prefer staying in their stage.
• Use the product strategy stack (mission, company strategy, product strategy, roadmap, goals) top-down to define direction, and bottom-up to debug missed targets.
• Make strategy concrete by sketching low-fidelity wireframes; they expose trade-offs and align teams far better than words alone.
• Set goals around understanding and execution when you lack a clear lever: categorize risk as understanding, dependency, execution, or strategic, and pick the matching goal type.
• Use the PM competencies framework to give specific, root-cause feedback, and ask your manager to rate you to kick off a concrete conversation.
• Practice selective micromanagement: step in tactically and temporarily to correct direction, then hand back autonomy once a clear framework exists.

02Frameworks mentioned

Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.