Monday Myth: Psychological Safety Is Not a Performance Strategy

A familiar phrase has taken hold across modern organisations: psychological safety.

It appears in leadership decks, hiring frameworks, team charters, and performance reviews. It carries a reassuring tone, suggesting responsibility and progress. The implied promise remains simple: create safety, unlock voice, and performance will follow.

Reality tends to resist that simplicity.

Psychological safety matters, but the belief that it drives performance on its own does not hold. When misunderstood, it produces the opposite effect and quietly undermines the very outcomes it intends to improve.

When Safety Becomes Comfort

In principle, psychological safety enables truth. In practice, it often protects comfort.

Difficult conversations soften, feedback loses precision, and underperformance gets reframed rather than addressed. Over time, teams learn where the invisible boundaries lie. They avoid friction and optimise for social acceptance instead of clarity.

The surface remains calm while the underlying structure weakens. Conflict does not disappear. It becomes suppressed. Problems persist, but their language fades, making them harder to confront and resolve.

From Safety to Moral Buffering

The drift reflects broader cultural patterns. In The Coddling of the American Mind, Lukianoff and Haidt describe how overprotection reshapes our relationship to discomfort. In The Righteous Mind, Jonathan Haidt explores how moral frameworks influence perceptions of harm, fairness, and disagreement.

Within organisations, this shift becomes operational. Safety moves from “freedom to speak truth” to “freedom from discomfort”. Disagreement begins to resemble harm, challenge appears as aggression, and direct feedback carries increasing social risk.

A specific mechanism accelerates this transition: the prioritisation of impact over intent.

On the surface, the idea appears reasonable. In practice, it alters behaviour. When perceived impact outweighs intent, precision becomes risky. Individuals hedge their language, soften their positions, and avoid stating what requires clarity. The cost of misinterpretation begins to outweigh the value of truth.

So‑called “safe spaces” can then invert their purpose. By narrowing what can be expressed without consequence, they restrict challenge and reduce the range of acceptable viewpoints. The intent remains protection. The effect becomes self‑censorship.

At scale, organisations shift from confronting reality to managing perception.

When Culture Initiatives Lose Their Edge

This pattern often emerges around inclusion and culture initiatives. Their original intent aims at fairness, stronger decisions, and broader perspectives, all of which remain valid.

The problem arises when the system drifts from outcomes to signalling.

Language begins to outweigh results. Disagreement turns into reputational risk. Challenge requires informal permission. Teams spend more effort avoiding missteps than solving problems.

At that point, the system no longer protects people so they can express truth. It protects them from the consequences of not expressing it.

Performance declines quietly.

A second effect compounds the issue. When belief systems, political or otherwise, enter the operating layer of the company, they displace the one criterion that must remain explicit and non-negotiable: performance. Organisations exist to deliver outcomes. Hiring, promotion, and evaluation anchored elsewhere introduce ambiguity and erode trust.

In practice, talent spreads widely, while opportunity does not. The role of the organisation consists in identifying and developing capability based on measurable contribution. As often highlighted by Jordan B. Peterson, capability distributions do not align neatly with social or political categories. Ignoring this reality leads to distorted signals.

When selection drifts toward signalling or ideological alignment, the organisation loses its ability to recognise and reward genuine capability.

The issue does not stem from diversity itself, but from any system that penalises challenge or substitutes performance with alignment to external beliefs. Such systems accumulate blind spots and degrade over time.

The Cost: Fragile Systems

The connection to Antifragile becomes difficult to ignore, along with a familiar dynamic akin to Goodhart’s law.

Systems that avoid stress do not gain robustness. They become fragile. When teams cannot sustain direct feedback, visible mistakes, or strong disagreement, they lose their capacity to adapt. Learning under pressure diminishes, replaced by a preference for stable perception.

At that stage, the proxy replaces the objective. Feeling safe replaces improving. Harmony replaces correction. As with Goodhart’s law, once safety becomes a measured target, it stops functioning as a useful signal and begins to distort behaviour.

When real pressure emerges, production incidents, deadlines, or market shifts, the system fails. The issue does not lie in competence, but in the absence of exposure to tension.

Safety Without Accountability

The distortion extends into ownership.

Frameworks such as Extreme Ownership (Jocko Willink) emphasise clarity of responsibility and accountability for outcomes. Misapplied safety weakens both.

Failures become collective, decisions diffuse, and responsibility spreads until it loses meaning. No individual remains clearly accountable, and improvement stalls.

Progress requires exposure. Something must prove wrong before it can improve. When systems protect individuals from being wrong, they also prevent the system from becoming right.

The same pattern appears in engineering practices. Pull request reviews shift from critical examination to polite validation. Postmortems turn into carefully constructed narratives that avoid tension instead of exposing root causes. Retrospectives favour agreement over correction.

In these environments, “safe spaces” extend into delivery practices. By limiting challenge, they reduce the system’s capacity to detect and correct failure.

What appears as maturity often reflects the absence of pressure.

What the Evidence Actually Suggests

Research on psychological safety, associated with Amy Edmondson and popularised by Google’s Project Aristotle, often suffers from oversimplification.

Psychological safety correlates with performance, but it does not cause it in isolation.

Project Aristotle identified five drivers of effective teams: psychological safety, dependability, structure and clarity, meaning, and impact. Safety ranked highly, yet only alongside clear expectations and reliable execution. Teams scoring high on safety but low on dependability or clarity did not outperform.

In Edmondson’s work, teams with high safety and low standards tend toward a “comfort zone”, while high standards with low safety create an “anxiety zone”. Only the combination produces a “learning zone”, where performance improves over time.

High-performing teams report higher safety because they trust competence, operate within clear standards, and deliver consistently. Safety emerges from a functioning system.

The key lies in balance:

High psychological safety combined with high accountability supports performance.

High safety without standards creates comfort. High standards without safety creates anxiety. Neither produces sustained results.

What Strong Teams Optimise For

Strong teams do not optimise for comfort. They optimise for truth.

They cultivate environments where feedback remains precise, disagreement remains acceptable, and mistakes surface early. They separate intent from impact without allowing either to dominate. They sustain tension long enough to resolve it.

Such environments demand both resilience and discipline. They require individuals capable of handling friction without escalation and correction without personalisation.

A More Honest Set of Questions

Most organisations ask the wrong question: whether people feel safe.

The question proves easy to measure and easy to present, yet provides little operational value.

• A more useful approach involves sharper probes.
• How quickly does a mistake become visible within the system?
• What follows when someone challenges a decision?
• Do incidents lead to corrective action or balanced narratives?
• Who can state that something is wrong without preparing a political defence?
• How often does truth prevail over consensus?
• Where does work slow because no one wishes to introduce tension?

These questions do not measure comfort. They reveal whether the organisation can confront reality without destabilising itself.

Closing

Psychological safety remains necessary, yet insufficient on its own.

When misunderstood, it shields organisations from discomfort. Progress, however, requires engagement with discomfort.

A team that feels safe yet fails to improve has redefined safety. It no longer enables truth. It protects from it.

That condition does not represent safety. It reflects drift.

High-performing organisations converge toward a demanding principle: environments where ideas face challenge without fear, and where growth depends on merit rather than alignment with prevailing narratives. Truth must outrun comfort, and contribution must outweigh conformity.

Without that balance, no degree of safety can compensate for the loss of reality.