Back to Blog

When User Interviews Lie: The Limits of Qualitative Research

By Jessica Chang

Early in my career, I made a critical mistake: I trusted user interviews too much. We spent weeks interviewing customers about a new feature. Everyone said they loved it. We built it exactly as discussed - and nobody used it.

The problem wasn't that users were deliberately lying. It's that humans are notoriously bad at predicting their own behavior and articulating their needs. We say we want healthy options, then order fries. We say we value privacy, then share personal details freely.

This disconnect between stated and actual behavior is why relying solely on interviews or surveys creates products that look good in presentations but fail in the market.

Better research combines methods: interviews tell you what people think, observation shows what they actually do, and analytics reveal patterns at scale. This triangulation gets you closer to truth.

When conducting interviews, focus less on hypotheticals ("Would you use this?") and more on past behavior ("Tell me about the last time you encountered this problem."). Specific stories reveal insights that general opinions can't.

Supplement interviews with observational methods. Usability tests, contextual inquiry, and diary studies all provide windows into actual behavior. Analytics and A/B testing validate findings across larger populations.

Finally, remember that users are experts in their problems, not in solutions. Your job isn't to implement what they request, but to uncover their underlying needs and design appropriate solutions.

The most valuable insights often come from resolving contradictions between what users say and what they do.