Respond productively to failure and bad news

How you react in the first thirty seconds after a mistake either banks or burns safety.

Why it works

Psychological safety is built and destroyed in the moments right after someone takes a risk. A productive response — curiosity about what happened, focus on the system not the person — reinforces that honesty pays. A punitive or shaming response teaches the whole watching team to hide the next problem until it is too big to hide.

How to do it

1. When something goes wrong, first ask what we can learn, before asking who is responsible.
2. Distinguish blameworthy failure (negligence) from intelligent failure (smart bets that didn’t pay off) and respond differently.
3. Visibly reward the surfacing of problems early, even unwelcome ones.

Evidence

Edmondson’s work on a "spectrum of reasons for failure" and on hospital error reporting shows that teams reporting more errors often had better, not worse, climates — they were surfacing problems, not making more of them. Her 1996 study of error detection first traced how group and organizational climate governs whether human errors get caught and corrected rather than hidden, and Cannon & Edmondson (2005) extended this to how organizations put failure to productive use. (observational)

Higher reported-error rates as a sign of safety is an inference from observational data, not a controlled comparison.

Sources

• Edmondson (2011), "Strategies for Learning from Failure", Harvard Business Review
• Edmondson, A. C. (2011). Strategies for Learning from Failure. Harvard Business Review, 89(4), 48–55.
• Cannon, M. D., & Edmondson, A. C. (2005). Failing to Learn and Learning to Fail (Intelligently): How Great Organizations Put Failure to Work to Innovate and Improve. Long Range Planning, 38(3), 299–319.
• Edmondson, A. C. (1996). Learning from Mistakes Is Easier Said Than Done: Group and Organizational Influences on the Detection and Correction of Human Error. Journal of Applied Behavioral Science, 32(1), 5–28.

Common mistake

Treating every failure identically — either punishing intelligent risks (which kills experimentation) or excusing genuine negligence (which kills accountability).