By Teresa Torres · producttalk.org · @ttorres on X · LinkedIn
Teresa Torres argues that customer interviews often fail because people describe their ideal behavior rather than what they actually do, and their brains readily invent coherent but false explanations. She draws on Michael Gazzaniga's split-brain experiments and Daniel Kahneman's work to show that confidence does not signal accuracy. She shares a costly product failure in passive-candidate recruiting that resulted from trusting a plausible story. Her remedy is to replace direct hypothetical questions with prompts asking for specific recent instances of behavior. This matters because product decisions built on idealized self-reports tend to churn customers and waste effort.
01Key takeaways
- Don't trust people's answers to hypothetical or general behavior questions; they tend to describe their ideal self.
- Ask customers to recount the last specific instance of a behavior rather than how they usually act.
- Treat confident, coherent explanations as possibly fabricated rather than as evidence of truth.
- Validate market hype and thought-leader claims against actual customer behavior before building.
- Follow up on whether a described instance is typical, and redirect generalizations back to specific stories.
- Compare your ideal process with your actual process to see where reality diverges from intention.
02Key sections
- The gap between stated and actual behavior
- A simple jeans-buying example shows that people state an ideal purchase factor but reveal different real drivers when asked about their last purchase. The same gap appears across everyday habits like gym visits and diet.
- Why our brains invent reasons
- Gazzaniga's split-brain study shows the left-brain interpreter fabricates plausible explanations for behavior it cannot know. Kahneman's point is that confidence reflects narrative coherence, not truth.
- What this means for product work
- Building from customers' ideal selves or from coherent but untrue stories leads to weak engagement and churn. Torres recounts a passive-candidate recruiting product that flopped because the team trusted a popular story over reality.
- Asking better interview questions
- Replace direct questions like 'what matters to you' with requests to tell about the last specific instance. Follow up with whether that instance was typical and avoid letting customers generalize with 'usually'.
- Try it yourself
- A short exercise compares an idealized process list for shipping a product with a list for the last real release. The two lists almost always differ, illustrating the same ideal-versus-reality gap.
03From the post
“I recently asked a woman what factors she considered when buying a new pair of jeans. She didn’t hesitate in her answer. She said, “Fit is my number one factor." That seems reasonable. It’s hard to find a pair of jeans that fit well. I then asked her”
04Frameworks mentioned
Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.