Be fundamentally different, not incrementally better

Jag Duggal, Chief Product Officer at Nubank (NewBank), discusses how the Latin American digital bank built a fanatical customer base and grows largely through word of mouth. The conversation covers product-market fit measurement with the Sean Ellis score, customer discovery, strategy versus execution, category design, and Nubank's plans for becoming a global, AI-driven personal banker.

01Key learnings

• Start from a deep, emotionally felt customer pain point rather than a feature idea; Nubank tied its origin to widely hated incumbent banks.
• Use the Sean Ellis survey as a gate before scaling: ask how disappointed users would be if the product disappeared, and raise the bar if your market is culturally polite or optimistic.
• Keep small, measure whether customers truly love it, and avoid scaling problems that aren't yet solved; scaling early creates big messes later.
• Be willing to say no and deliver bad news early; owners, not renters, flag doomed projects before pressure makes it costly.
• Talk to customers yourself by phone rather than relying on layers of research summaries, and listen for tone and nuance that statistics miss.
• Write a crisp hypothesis before research and stay open to disproving it; observe behavior more than asking direct questions.
• Strategy should be specific: a coherent diagnosis, a focused bet, and a clear advantage, rather than a vague vision or ambitious goal. Being clear lets you detect when you are off course.
• Aim to be fundamentally different, not incrementally better, and consider defining or reinventing a category when you enter a market.

02Frameworks mentioned

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