Kezia knows what she needs to say.
She is sitting in the weekly leadership meeting, third seat from the end of the table, the one she always takes because it is close enough to be visible but far enough from the center to feel safe. The new product is about to launch. She has identified a risk — a real one, not a theoretical one — in the rollout timeline. She has the data. She has thought through the implications. She has even drafted a note to herself about how she would raise it.
She almost raises her hand.
Then she remembers what happened to Marcus, eight months ago, when he flagged something in a similar meeting. She remembers the micro-pause before the senior VP responded. The way the conversation moved past his concern as though it had not quite been heard. The way the project moved forward anyway. The way Marcus went quiet in meetings after that — not suddenly, not dramatically, just gradually, incrementally, permanently quieter.
Kezia does not raise her hand.
Two months later, the risk she identified materializes. The launch is delayed. The team scrambles. The senior VP sends an email about the importance of surfacing concerns early.
No one in the room misses the irony.
The Real Definition of Psychological Safety
Amy Edmondson’s research at Harvard Business School has given organizations a framework that is now nearly ubiquitous: psychological safety is the belief that one will not be punished or humiliated for speaking up with ideas, questions, concerns, or mistakes.
Most organizations have read about it. Many have launched initiatives around it. Some have even put it in their values documentation.
And yet Kezia still does not raise her hand.
Here is what I think gets missed in the standard framing: Edmondson’s research is not primarily about creating a comfortable environment. It is about creating a predictable one. Psychological safety, at its core, is about people knowing what will happen when they take the risk of honesty.
Not what leadership promises will happen. Not what the values document says will happen.
What has actually happened. What the evidence — gathered through years of careful, often painful observation — says happens in this specific room, with these specific people, when someone takes the risk.
Kezia was not irrational. She was accurate.
The Narrative Infrastructure No One Is Auditing
Every organization has what I call a narrative infrastructure of safety. This is the web of stories that circulates through an organization — not the official communication, not the policies, not the training — about what actually happens when someone speaks up.
These stories travel quickly and hold long. They are told in hushed voices over lunch, in quick hallway exchanges, in the subtext of what gets brought to meetings and what gets held back. They are updated every time something happens that confirms or challenges them. They are absorbed by new team members within months of joining, sometimes within weeks.
And they are more powerful than any policy you will ever write.
When I work with organizations on their culture, I often ask leaders what story is currently circulating about what happens when someone challenges a decision in this company. The question produces a significant pause. Not because leaders do not know the answer — most do, if they are honest with themselves — but because most have never been asked to name it directly.
The narrative infrastructure is the real policy. It is the one that governs actual behavior.
Four Questions to Audit Yours
The work of building psychological safety is not the work of writing a new values statement or redesigning an all-hands meeting. It is the work of surfacing and changing the narrative infrastructure — the stories people are already telling, right now, about whether honesty is safe here.
These four questions are designed to help leadership teams do exactly that.
What happened to the last person who raised a difficult concern? Not in general. Specifically. Think back to the most recent instance where someone challenged a decision, flagged a risk, or pushed back on direction in your organization. What happened to them in the aftermath? Did they get credit? Did they get passed over for something? Were they thanked publicly, or quietly sidelined? The answer to this question is the current definition of psychological safety in your organization — not the official one, the real one.
What story is circulating about vulnerability at the leadership level? Leaders model what is permissible. When leaders share their uncertainty, acknowledge their mistakes, and name their emotional reality in organizational contexts, they expand the range of what team members understand to be allowed. When leaders present consistent polish and certainty, they communicate — clearly, if unintentionally — that vulnerability is not safe here. What has your leadership team modeled, publicly and recently, about what it looks like to be human at work?
When was the last time leadership was wrong — and said so? This is different from accepting feedback gracefully in one-on-ones. This is about public acknowledgment. Did leadership ever communicate, to the wider team, that a decision turned out to be wrong, and here is what was learned? The absence of this story does not signal infallibility. It signals that error is not survivable. And in organizations where error does not feel survivable, people stop taking risks — including the risk of honest communication.
What do people protect themselves from when they hold back? This last question is the most revealing. When someone in your organization chooses not to raise a concern, what are they protecting themselves from? Embarrassment? Exclusion? Being labeled as a problem? Being subtly disadvantaged in the next round of decisions? The answer tells you what the narrative infrastructure currently rewards — and what it punishes.
Changing the Story
Organizations often attempt to build psychological safety by changing the visible layer: new values, new policies, new training programs, new language from HR about creating an inclusive environment.
These things are not useless. But they are surface. They do not touch the narrative infrastructure. And the narrative infrastructure is what Kezia is using to decide whether to raise her hand.
The real work is slower and more specific. It is an act of leadership — not program management. It requires leaders to be intentional about creating evidence. Evidence that honesty has a future here. Evidence that vulnerability is not a career risk. Evidence that the story about what happens when you speak up is being rewritten, one visible, deliberate, consistent act at a time.
This is the work I do through Narrative Business Practice™. Not changing the policy. Changing the story that determines whether the policy is believed.
The Story That Has to Come First
If you are a leader reading this and you want your team to take the risk of honesty, there is a story you have to tell first.
Not the story about your open door policy.
The story about the last time you were wrong, and what it felt like, and what you learned, and why you are still here talking about it.
That story — the true one, the one that costs something to tell — is the only story that changes Kezia’s calculation.
She is waiting to hear it.
