CHAPTERS
Why “good and never-been-thought-of” is a trap
Dalton explains that most people unconsciously filter startup ideas in similar ways. If your goal is to find an idea that is both obviously good and completely novel, you’ll likely fail because the search space is unrealistically constrained.
Flip the search: pursue ideas others discarded
Instead of hunting for an untouched concept, Dalton suggests looking for ideas many people have considered and rejected. The existence of prior rejection can be useful data—if you can understand why it was rejected and what’s changed (or what others missed).
“Find what everyone thinks is bad” as a uniqueness heuristic
Dalton reframes uniqueness as moving toward ideas that trigger skepticism. If many people think an idea is bad, it may indicate you’re outside consensus—where outsized startup returns often live.
Instacart as an example of a dismissed idea that worked
Michael cites Instacart as a classic case of an idea that many would have written off. The point is that “sounds bad” or “already tried” doesn’t mean “can’t win,” especially when conditions or execution improve.
Avoid ideation-by-friend-group approval
Michael warns that founders often validate ideas by asking friends and looking for universal enthusiasm. He suggests the opposite: friend approval may simply reflect shared biases and a consensus-seeking filter.
Look for “controversial,” not universally liked
Michael clarifies the target isn’t an idea everyone hates, but one that creates disagreement or concern. If some people respond with “I’m worried about you,” it can be evidence you’re exploring territory outside the default playbook.
Airbnb-style skepticism: friends and family think it’s a bad idea
Dalton and Michael reference the common pattern among successful YC companies: friends and family doubting the founders’ idea. The broader lesson is that early social disapproval is normal when an idea is genuinely different.
A practical way to escape consensus: pick hard, long-horizon problems
Michael offers another framework for moving away from consensus: choose things that are difficult and take a long time. He argues founders often self-limit by insisting something will work in two years, eliminating many potentially higher-upside ideas.
Get more out of YouTube videos.
High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.
Add to Chrome