By Teresa Torres · producttalk.org · @ttorres on X · LinkedIn
Teresa Torres argues that product builders consistently overestimate how right they are, drawing on behavioral research and the Heath brothers' book Decisive, which describes four decision-making traps: narrow framing, confirmation bias, short-term emotion, and overconfidence. Because we rarely feel uncertain, we keep building on flawed assumptions. Her remedy is to expect to be wrong, define measurable outcomes and explicit hypotheses in advance, and build systems that catch errors early so teams can course correct quickly. For product work, the core skill is not having big ideas but detecting failure fast.
01Key takeaways
- Assume your product assumptions are probably wrong, and treat confidence as a warning sign rather than evidence.
- Frame decisions as how we can do this and that, not only whether to do this or that.
- Write down hypotheses and success thresholds before building so results cannot be reinterpreted later.
- Actively seek evidence that disconfirms your beliefs instead of only collecting supporting data.
- Set up measurement and review rhythms that reveal failing bets early enough to correct course.
02Key sections
- Our confident answers
- Product people readily produce answers about features and user behavior, and those answers feel right, which is precisely the problem. Research shows our minds rarely register being stumped.
- Four decision-making villains
- Drawing on Decisive, the post names narrow problem framing, confirmation bias, short-term emotion, and overconfidence as the main ways decisions go wrong.
- Revising history
- After the fact, people tend to reinterpret results as what they expected, which hides the signals that something was not working.
- Building systems to catch being wrong
- Good product work depends on mechanisms that surface errors early, so teams can adjust course often rather than occasionally.
03From the post
“This post is part of a series about making better product decisions. As startup founders, product managers, designers, it's often easy for us to come up with all the answers. Should we add this feature? Will users take this action? Will changing this copy increase conversions? And of course,”
04Frameworks mentioned
Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.