Psychological safety has become one of the workplace’s most fashionable promises, with all sorts of management consultants making these unfounded promises.
We are told that psychologically safe teams speak up more, learn faster, innovate better and make fewer mistakes. I have little argument with the destination. My problem is with the vehicle.
Ironically when one speaks up and asks where is the test data from the operating system? Cue the crickets.
If an organisation were promoting a medical treatment, (like a revolutionary blood test), we would ask how it had been tested, what the control group showed, and whether the treatment itself—not merely the desired outcome—produced measurable results.
Yet psychological safety is often promoted differently. We measure whether people feel safe, correlate that feeling with desirable outcomes, train leaders to model certain respectful behaviours, and then talk as though we possess a tested mechanism for reliably producing safety.
I think that leap deserves scrutiny.
I think we should be asking for the equivalent of a clinical trial: a clearly specified behavioural process, used consistently across teams, with defined interventions, escalation rules, measurable outcomes and replicated results. I suspect that’s when the evidence cupboard starts looking remarkably empty.
That does not make claims of psychological safety necessarily fraudulent. But promoting its benefits while remaining vague about the mechanism for producing it can begin to resemble the Emperor’s new clothes: everyone admires the garment, while few ask what is there.
My interest is not in someone promising psychological safety. It is finding and understanding the operating system underneath it.
What happens when Jim talks down to Sue, after she spoke up sharing her latest concept?
What can Sue do, in real time?
Should she toughen up, and empathise with Jim, imagining that he probably had a bad day?
Maybe she should just appreciate his feedback, regardless.
What is Jim required to do next?
What if he ignores her and continues being uncivil?
When does the issue escalate, who reviews it, and what gets recorded?
How do they make peace after?
And finally where is the tested solutions and real-time data promised?
Until we can answer those questions operationally, “psychological safety” risks remaining an aspiration dressed as a some effective system.
Safe to say, I think we need to test the clothes or expose that they don’t exist.
Leave a Reply