A happy team and a psychologically safe team are not the same thing. I think confusing them is how good managers end up avoiding hard conversations. A new essay on charity.wtf, part of a dialogue with Dr. Cat Hicks, argues that psychological safety drifted into a promise that nobody feels bad. Its real purpose was to make learning possible.
The essay says what safety is not. For a manager, I think the more useful question is how you would see it. A mood survey will not tell you. How early bad news arrives will tell you. Look for an estimate flagged as slipping early in the sprint instead of the day before the demo, a reviewer who says the approach is wrong before the merge, or a developer who admits they do not fully understand an AI-generated change.
The essay notes that the effect is strongest in uncertain, interdependent work. Web teams adopting coding agents are doing exactly that kind of work right now.
In fifteen years of running web teams, I have come to believe a team can be content and still keep every problem from its manager. I have also come to believe that uncomfortable conversations are usually the evidence that safety exists.
I would like to hear what other leads watch for, apart from a survey, that tells them people feel able to bring bad news.
Photo source: https://photos.robertstowe.com/bengaluru