Build empathy before committing to a solution
⏱ Untimed topicBelsky says teams can become passionate about their preferred answer while missing the customer's actual problem. Understanding the person suffering the problem may lead to a different solution. Product sense starts with that understanding rather than only confident taste.
Make the first moments feel successful
⏱ Untimed topicHis lazy, vain and selfish shorthand describes new users wanting quick progress, a good impression and early success. It is a product-design lens, not a universal psychological diagnosis. Reduce the learning burden before asking a user to care deeply about your mission.
Treat orientation and defaults as onboarding
⏱ Untimed topicHe defines the first mile as more than a tour or signup sequence. It includes knowing where you are, how to get home, where to find help and what defaults appear. A functional flow can still leave a new person lost.
Revisit onboarding as the audience changes
⏱ Untimed topicEarly customers may tolerate more friction than later, more skeptical customers. Belsky says a once-effective introduction may stop working for a new cohort. Review the experience for those arriving now instead of assuming past onboarding results settle it.
Optimize for the problems you want to have
⏱ Untimed topicHe prefers customers reaching value and then asking for more capabilities over a broad launch that prevents them reaching value at all. For an MVP, prioritize obstacles on the path to caring. This is sequencing advice, not a reason to ignore safety or essential customer needs.
Remove brick walls and distracting options
⏱ Untimed topicHe names signup, account connection and login as potential blockers. His Behance retrospective describes removing secondary features and seeing more core publishing activity. Those effects are his account, unverified here, and removal should be tested rather than assumed to help every product.
Preserve time to explore before narrowing
⏱ Untimed topicBelsky describes strong designers and product leaders exploring many possibilities, selecting a smaller set, getting feedback and refining further. The team then sees a few considered options rather than every first idea. This selected point does not cover his later generative-AI discussion.
