What You're Actually Trading When You Ask for Psychological Safety

In 1999, Amy Edmondson set out to study whether better-performing nursing teams made fewer errors. She expected they would. They didn’t. High-performing teams reported significantly more mistakes than their lower-performing counterparts - not because they were less careful, but because they felt safe enough to admit them. The worse teams looked cleaner on paper. They were just quieter about what went wrong.

It’s the kind of finding that inverts the intuition. And it sits underneath most of what we misunderstand about psychological safety.

What it actually is

Psychological safety isn’t a culture of niceness. It isn’t the absence of conflict, or a promise that nothing uncomfortable will be said. Edmondson’s definition is more precise: it’s the shared belief that the team is safe for interpersonal risk-taking. The permission to be wrong, to ask the question that might sound obvious, to raise the concern before it becomes a crisis.

The distinction matters because a lot of teams optimise for the appearance of it - the open-door policy, the “no bad ideas” framing in the kickoff - without creating the conditions where people actually use it.

Why it isn’t universal

Google’s Project Aristotle (2016), which analysed over 180 internal teams, found psychological safety was the strongest single predictor of team effectiveness. It’s a compelling finding. But it’s worth reading carefully.

The teams in that study were largely working on complex, ambiguous problems - the kind where surfacing doubt early and challenging assumptions has a direct effect on outcomes. The finding doesn’t necessarily generalise to every role, every team, or every kind of work.

For someone in a creative or iterative role - a product team, a research function, a group navigating genuine uncertainty - low psychological safety is a structural problem. It breaks the feedback loop the work depends on. People make decisions with incomplete information because no one’s comfortable saying what they actually think.

For someone in a compliance or operations role, where the work is built around consistency and established procedure, the calculus is different. Speaking up still matters. But psychological safety isn’t the load-bearing wall of that team’s performance in the same way.

When it’s missing

The absence of psychological safety rarely announces itself. What you notice instead is a meeting where the hard question doesn’t get asked. A post-mortem where everyone agrees on a tidy version of what went wrong. A team that seems aligned - and probably is, on the surface.

What’s harder to see is what that costs over time: the small problems that compound quietly because surfacing them felt riskier than ignoring them.

When it’s there

A psychologically safe environment doesn’t feel frictionless. If anything, it feels slightly more uncomfortable - because disagreement is audible, uncertainty gets named, and someone always asks the question everyone else decided not to ask.

Edmondson describes it not as comfort but as a particular kind of courage. Not the individual courage to speak up regardless of consequence, but a collective decision that speaking up is part of how this team operates. That framing shifts something: it’s not a personality trait, it’s a property of the relationship.

The tradeoff

Psychological safety costs friction. More of it, and more visible. Teams with high psychological safety move more slowly than teams where people just get on with it. Tension comes into the room rather than staying outside the door. For a team where iteration and honest feedback are the whole point, that friction is necessary and valuable. But it is still friction.

There’s something more personal in the tradeoff, too. Most people who say they want a psychologically safe environment have tested that preference in one direction: whether they’d feel comfortable speaking up. Fewer have tested it in the other - whether they’re equally comfortable when the candour is directed at them. At something they did, a decision they made, a view they hold.

Plenty of people want transparency in principle. It’s a different thing in practice.

Where this sits in your hierarchy

Psychological safety is one of 19 factors in Kanso’s assessment, sitting under Cultural Fit. The tool doesn’t assume it’s a must-have. It asks you to decide whether it is - and what you’d trade to have it, or give up by not having it.

To find out where it sits in your own hierarchy: https://trykanso.app

Have you ever worked in a team where psychological safety genuinely worked - and was the version of it you got the version you thought you wanted?