Psychological Safety Audit¶

Why It Matters¶

When psychological safety is low, people engage in defensive self-protection: hiding mistakes until they're catastrophic, not asking questions that reveal knowledge gaps, withholding dissenting information that could prevent bad decisions, and avoiding experiments that might fail. This protection is individually rational but collectively catastrophic. Organizations with low psychological safety appear to have incompetent, uncurious, conformist employees. In reality, they have normal people behaving rationally in unsafe environments. When psychological safety is high, the same people ask "dumb" questions that reveal critical gaps, flag problems early when they're fixable, and propose ideas that might fail but might revolutionize. The environment, not the people, determines the behavior.

The Key Move¶

To assess psychological safety, check for specific behaviors: Do people admit mistakes before they escalate? Do junior people challenge senior people's ideas? Do people ask clarifying questions that reveal they don't understand? Do experiments get run where failure is plausible? Do meetings have genuine disagreement or performative consensus? If these behaviors are rare, psychological safety is low. To improve it, leaders must visibly model: admitting their own mistakes, rewarding people who flag problems, responding to challenges with curiosity not defensiveness, and explicitly appreciating risk-taking even when it fails.

Classic Application¶

Google's Project Aristotle research on team effectiveness. After analyzing hundreds of teams, the strongest predictor of team performance wasn't talent, resources, or structure—it was psychological safety. Teams where members felt safe taking interpersonal risks consistently outperformed teams with superior individual talent but lower safety. Why? High-safety teams surfaced problems early, learned from failures, combined diverse perspectives, and adapted quickly. Low-safety teams appeared harmonious but suppressed dissent, hid problems, and repeated mistakes. The insight: safety is a structural precondition for performance, not a soft "nice to have."

Surprising Application¶

Code review culture in software teams. Some teams have high-quality, rigorous code reviews where reviewers catch bugs, suggest improvements, and submitters learn. Other teams have perfunctory reviews where everything gets "LGTM" approval despite obvious issues. The difference is rarely reviewer competence—it's psychological safety. In low-safety environments, thorough reviewing feels like personal attack; submitters get defensive; reviewers learn to stay superficial to avoid conflict. In high-safety environments, thorough review is collaborative improvement; submitters appreciate catches; reviewers feel safe being direct. Same practice, opposite outcomes based on safety. Building safety requires modeling: senior engineers having their code substantively reviewed, thanking reviewers for catching issues, and treating review comments as learning not judgment.

Failure Modes¶

Confusing safety with comfort—psychological safety means safety to take risks and disagree, not protection from all negative feedback. Attempting to mandate safety—you can't force people to feel safe; it must be built through consistent leadership behavior. Assuming safety eliminates all conflict—healthy conflict increases with safety as people voice disagreements rather than suppressing them. Using "lack of safety" to excuse avoiding difficult conversations—sometimes people need clear feedback even if uncomfortable. Measuring safety through surveys alone—actual behaviors (dissent, error admission, help-seeking) are better indicators than stated feelings.

Go Deeper¶

Edmondson, "Psychological Safety and Learning Behavior in Work Teams," Administrative Science Quarterly 44(2), 1999; Edmondson, The Fearless Organization (2018).