Week 15 Day 2: What a Trust Audit Actually Looks Like

Surveys measure what people are willing to write down. Conversations measure what people are willing to say. A Trust Audit uses one-on-one conversations with a consistent set of questions to build a trust profile of your team. It takes about 20 minutes per person. It surfaces information that no survey, no engagement score, and no pulse check will ever capture. And it works because the act of asking itself demonstrates the trust you are trying to build.

Here is the format. Schedule a 20-minute one-on-one with each team member. Open with: 'I am running a trust audit -- a set of questions I am asking everyone on the team to understand how we are doing on trust, candor, and safety. There are no right answers. I am looking for honesty, not diplomacy.' Then ask these five questions. One: 'On a scale of 1 to 10, how comfortable are you disagreeing with me in a meeting?' Two: 'When was the last time you held back an opinion because you were not sure how it would be received?' Three: 'If you made a significant mistake tomorrow, what would your first instinct be -- tell me immediately, fix it quietly, or tell a peer first?' Four: 'Is there anything about how this team operates that you have wanted to change but have not felt safe bringing up?' Five: 'What is one thing I do as a leader that makes trust harder?' Do not defend yourself. Do not explain. Just listen, take notes, and say 'thank you.' The most important thing you learn will come from the question that makes you most uncomfortable.

The Trust Audit format is adapted from Lencioni's (2002) team health assessment in 'The Five Dysfunctions of a Team,' combined with elements of Edmondson's (1999) psychological safety measurement protocol. Lencioni's model identifies absence of trust as the foundational dysfunction, while Edmondson's research provides specific behavioral indicators (willingness to take interpersonal risks, asking for help, admitting mistakes) that operationalize trust into observable actions. The one-on-one format rather than group survey is deliberate: research by Totterdell, Kellett, Teuchmann, and Briner (1998) on emotional contagion in work groups demonstrates that group settings amplify both positive and negative sentiment, making individual responses less accurate. The five-question structure follows what Patton (2015) in 'Qualitative Research and Evaluation Methods' calls 'standardized open-ended interviews' -- consistent questions that allow comparison across respondents while preserving the richness of individual responses. Question five ('What do I do that makes trust harder?') is the most diagnostic because it inverts the typical feedback dynamic: instead of asking what the team should change, it asks what the leader should change. This inversion leverages what Schein (2013) calls 'humble inquiry' and what Edmondson (2019) identifies as the single most powerful psychological safety intervention: a leader explicitly asking for feedback on their own behavior.