The Research on Psychological Safety
Psychological safety in your team may be the single most important factor determining whether it thrives or underperforms — and Google's landmark Project Aristotle proved exactly that. After studying hundreds of teams to identify what made some consistently outperform others, researchers found that the biggest predictor of team effectiveness was not individual talent, experience, or even organizational structure. It was whether team members felt safe enough to take interpersonal risks — and the teams that did consistently left the rest behind.
What Psychological Safety Is and Is Not
Psychological safety is not about being nice all the time or avoiding hard conversations. It is about creating an environment where people believe they will not be punished or humiliated for speaking up, asking questions, making mistakes, or challenging the status quo.
How Leaders Build Psychological Safety
- Model vulnerability: Share your own uncertainties and mistakes openly—this signals that it is safe to do so.
- Respond to bad news without shooting the messenger: How leaders respond to problems determines whether they hear about them early or late.
- Actively invite dissent: Ask explicitly for challenge and counterarguments in discussions.
- Follow through on concerns raised: When someone raises a concern and nothing happens, it teaches others not to bother.
- Give credit generously and take blame proportionally: The inverse is what destroys safety fastest.
The Role of Consistency
Trust is built slowly and destroyed quickly. Psychological safety requires consistent behavior from leaders over time. One instance of publicly shaming a team member for a mistake can undo months of trust-building. Leaders must be vigilant about the signals they send—both in what they do and in what they fail to do.
The Business Case for Psychological Safety
Beyond performance, psychological safety directly impacts retention, innovation, and the speed at which problems get surfaced and solved. Organizations with psychologically safe cultures catch errors earlier, iterate faster, and retain their best people longer. It is not a soft initiative—it is a competitive advantage.
Measuring Psychological Safety on Your Team
You cannot improve what you do not measure, and psychological safety is no exception. The most widely used diagnostic approach draws on Amy Edmondson's original research, which produced a short set of perception-based survey items asking team members to rate how comfortable they feel speaking up, admitting mistakes, and disagreeing with the majority. Administering even a condensed version of this instrument on a quarterly basis gives leaders a reliable baseline and a way to track whether their interventions are actually moving the needle.
Quantitative surveys tell you where you stand, but qualitative signals often tell you why. Pay close attention to meeting dynamics: who speaks first, who stays silent, how quickly consensus forms, and whether the same voices dominate every conversation. A team where only senior members contribute ideas or where meetings end suspiciously fast is likely operating below the safety threshold, even if survey scores look acceptable on paper.
Leaders should also track proxy metrics that correlate with psychological safety over time, such as the ratio of problems surfaced proactively versus discovered reactively, voluntary turnover among high performers, and the frequency of cross-functional feedback. None of these metrics is a perfect proxy on its own, but together they paint a picture of whether people feel genuinely secure enough to engage fully with their work and with each other.
Common Leader Behaviors That Erode Safety
Many leaders damage psychological safety not through dramatic acts of aggression but through subtle, repeated behaviors they are often unaware of. Interrupting team members mid-sentence, visibly checking a phone during someone's presentation, or responding to a novel idea with an immediate list of reasons it will not work all send a clear signal: your input is not valued here. Because these moments accumulate quietly, leaders rarely receive direct feedback about them until the damage is already done.
Inconsistency is one of the most corrosive forces a leader can unleash. When certain individuals are allowed to bypass processes, escape accountability for poor outcomes, or speak over colleagues without consequence, the rest of the team draws its own conclusions about whose voices actually matter. Favoritism — even when entirely unintentional — teaches people that safety is conditional, and conditional safety is functionally the same as no safety at all.
A particularly dangerous pattern is the performative invitation to speak up, where a leader asks for honest input and then visibly dismisses or penalizes the first person who provides it. This single exchange can silence a team for months. Leaders who want to diagnose whether they have fallen into this trap should ask themselves honestly: when was the last time someone on my team told me something I did not want to hear, and how did I respond in the room in that moment?
Psychological Safety in Remote and Hybrid Teams
Remote and hybrid environments introduce structural friction that makes psychological safety harder to build and easier to lose. The informal exchanges that naturally occur in a shared physical space — the hallway conversation, the offhand comment over coffee — are precisely the low-stakes interactions that gradually lower people's guard and deepen trust. When those moments disappear, leaders must be deliberate about creating equivalent opportunities in digital environments, because they will not emerge on their own.
Video calls create a specific set of dynamics that leaders need to actively manage. The default tendency for participants to self-censor increases when interactions feel transactional or when the meeting format makes it socially awkward to interrupt or push back. Structuring time for open reflection, using asynchronous channels for complex questions that benefit from thoughtful responses, and explicitly rotating who leads or facilitates discussions can help distribute voice more equitably across a distributed team.
For hybrid teams specifically, the risk is that psychological safety becomes unevenly distributed — higher for those physically present and lower for those joining remotely. Leaders who consistently give in-person attendees first opportunity to speak, make decisions in informal hallway conversations, or fail to ensure that remote participants have equal visibility into context and information are inadvertently building a two-tiered culture. Closing that gap requires both structural changes to how meetings are run and a genuine commitment to treating digital presence as equivalent to physical presence.
Rebuilding Safety After a Trust Breach
When a significant trust breach occurs — a public reprimand, a broken confidence, a decision made without the team's input that directly affects them — the path back is neither quick nor automatic. The first and most critical step is a direct, unambiguous acknowledgment from the leader, not a softened corporate-style apology that hedges responsibility, but a clear recognition of what happened, why it was harmful, and what specifically will change. Teams are remarkably attuned to whether an apology is genuine or performative, and a hollow one accelerates the erosion rather than reversing it.
After the acknowledgment, leaders must accept that trust will be rebuilt through behavior over time, not through words or a single gesture. This means being exceptionally consistent in the period following a breach — perhaps more so than at any other point in the team's history. Every interaction during this window is being evaluated against the implicit question: has anything actually changed? Responding well to even small moments of vulnerability or dissent during this period signals more than any formal statement could.
It is also worth addressing the breach openly with the broader team when appropriate, rather than assuming that because something was not discussed, it was not noticed. In most cases, team members are acutely aware of what happened and are watching to see how leadership handles it. Bringing the issue into the open, framing what was learned from it, and inviting dialogue about how the team moves forward can actually strengthen psychological safety beyond its pre-breach level when handled with genuine care and humility.
Psychological Safety at Scale Across Departments
Building psychological safety within a single team is a meaningful achievement, but senior technology leaders and CIOs face a more complex challenge: creating the conditions for it to take root across an entire organization or multiple departments simultaneously. At scale, the leader's direct influence over day-to-day team dynamics diminishes significantly, which means the work shifts from personal modeling to system design. This includes how performance is evaluated, how mistakes are documented and discussed, how cross-functional conflict is resolved, and whether the organizational structure itself rewards candor or punishes it.
One of the most powerful levers available to senior leaders is the behavior they visibly expect from the managers who report to them. If a CIO holds their direct reports accountable for the psychological safety of their own teams — asking about it in one-on-ones, reviewing it alongside other team health indicators, and recognizing leaders who cultivate it — the norm cascades downward through the organization. Conversely, if psychological safety is treated as a values statement on a wall rather than a measurable leadership responsibility, middle managers receive no real signal that it matters.
Cross-departmental psychological safety adds another dimension: whether people feel safe enough to challenge ideas, flag risks, or ask for help across team boundaries, not just within them. Siloed organizations often have pockets of internal safety that do not extend across functional lines, leading to the suppression of exactly the kind of cross-domain insight that drives innovation. Leaders who want to close this gap should invest in creating shared forums where candor across departments is explicitly modeled from the top, and where early signals of inter-team tension are treated as important data rather than inconvenient noise.