What Actually Builds Psychological Safety at Work
"Psychological safety" gets used a lot in workplaces, often printed on a values page rather than something anyone's actually built. It's worth being specific about what this concept actually refers to, what research broadly suggests tends to build it, and, from my own experience, what it can feel like when it's genuinely missing.
What the concept actually means
Psychological safety, broadly, refers to a shared belief within a team that it's safe to take interpersonal risks, asking a question, admitting a mistake, challenging an idea, without fear of embarrassment or being penalised for it. Some organisational research has suggested this shared sense of safety predicts a team's willingness to engage in learning behaviours, feedback-seeking, experimenting, discussing errors openly, and that it's this openness, more than raw skill or individual talent, that tends to predict better performance over time. Separately, workplace research into what distinguishes high- and low-performing teams has pointed in a similar direction, suggesting psychological safety may matter more than experience or technical ability alone.
One theme that comes up in this area is worth sitting with, because it's a little counterintuitive: some research suggests that higher-performing teams don't necessarily make fewer mistakes than lower-performing ones, they may simply feel safer surfacing problems rather than quietly managing around them.
What this can feel like when it isn't there
A couple of examples of what this might look like, though these are illustrative rather than the full picture. In previous roles, I've noticed two quite different things happen when I said something honestly. One was sharing something in what I understood to be confidence, only to notice it circulating afterward, further than I'd ever intended it to go, less like the private conversation I thought I was having and more like something that had quietly become material for other people to discuss. The other was saying something and feeling dismissed or minimised, like the thing I'd raised had been quietly filed away rather than actually taken in.
What stayed with me from both was the same instinct afterward: pulling back, saying less, testing the water less, becoming more careful about what got shared and with whom, and where it might end up once it left my hands. Looking back, that pulling back seems to line up with the general theme in this area, not disengagement out of laziness or disinterest, but a fairly reasonable response to an environment that hadn't made room for the risk of speaking plainly, or hadn't held what was shared as carefully as I'd assumed it would be.
What this suggests, practically, as a coach
Normalise naming mistakes early, rather than treating error-reporting as something to minimise or bury. Teams that hide problems tend to solve them later, and worse, once they've had time to compound.
Ask more than you tell. Genuine, open questions signal that other perspectives are actually wanted, not just tolerated as a formality.
Respond to bad news without visible punishment, even subtle punishment, a sharper tone, a slight withdrawal. How a leader reacts the first time someone flags a problem tends to set the pattern for whether it happens again, often more than any stated policy does.
Separate the person from the mistake, explicitly and out loud. This sounds small, but it's often the actual mechanism that lets people keep taking interpersonal risks rather than quietly stop.
Notice what happens to information after it's shared. If something said in confidence, or in good faith, seems to circulate further afterward, that's often what teaches people to stop sharing, faster than almost anything else.
Build it deliberately, rather than assuming it exists because it's been named. Psychological safety tends to be something a team has to actively construct and maintain, not a by-product of good intentions or a values statement.
Where this leaves things
If a team says the right things about safety but people still hesitate to speak up, or quietly stop after one bad experience, the safety probably isn't actually there yet, whatever the values statement says. It's something built through repeated, consistent behaviour, especially in how information and mistakes get handled, not declared into existence by naming it. I know, from my own experience of it being absent, how much that gap between the stated value and the lived reality can cost, in confidence, in openness, and in what people are willing to bring to a team at all.