In the search for what makes teams effective, organizations have invested heavily in hiring the right people, building complementary skill sets, and establishing clear goals. These factors matter. But over the past two decades, a growing body of research has converged on a less obvious variable, one that research increasingly links to how well teams learn and perform. That variable is psychological safety: the shared belief among team members that the team is safe for interpersonal risk-taking.
The concept has moved from academic journals to boardroom conversations, in large part because of Amy Edmondson's foundational research at Harvard and Google's widely publicized Project Aristotle. Yet despite its visibility, psychological safety remains one of the most misunderstood constructs in organizational life. It is not about being nice. It is not about avoiding conflict. And it is not something that can be created by a single workshop or policy memo. Understanding what the research actually says, and what it means for leadership behavior, is essential for any organization serious about building high-performing teams.
Free · 3 minutes
How safe is it to speak up around you?
Take the Free Psychological Safety Check: 12 quick choices, four habits a leader controls, and your result on screen with no email needed.
What Does the Research Actually Show?
Amy Edmondson first defined team psychological safety in her landmark 1999 paper as "a shared belief held by members of a team that the team is safe for interpersonal risk-taking" (Edmondson, 1999). This means team members feel confident that they will not be punished, humiliated, or marginalized for speaking up with questions, concerns, mistakes, or half-formed ideas. Her earlier study of hospital units found that better-led units reported more medication errors, not fewer, likely because people there felt safe to report them (Edmondson, 1996). Her 1999 study of 51 teams in a manufacturing company then showed that psychological safety predicted team learning behavior, which in turn predicted team performance. Psychologically safe teams surfaced problems early, learned from them, and adapted. Teams without it were more likely to hide mistakes and avoid hard conversations.
What made this finding so striking was its counterintuitive nature. Leaders who assumed that fewer reported errors meant better performance were systematically rewarding the wrong teams. The quiet, seemingly smooth-running teams were often the ones most at risk, because silence was masking problems rather than reflecting their absence.
Edmondson's work has since been replicated and extended across industries, cultures, and organizational levels. A meta-analysis by Frazier, Fainshmidt, Klinger, Pezeshkan, and Vracheva (2017) pooled 136 samples covering more than 22,000 people and found that psychological safety was related to information sharing, learning behavior, organizational commitment, job satisfaction, and task performance.
Perhaps most compelling to business audiences was Google's Project Aristotle, an internal study (not a peer-reviewed one) of about 180 teams across the company that looked for what distinguished high-performing teams from average ones (Duhigg, 2016). Google's researchers concluded that who was on a team mattered far less than how the team worked together. The single most important factor was psychological safety. It was the foundation on which the other four dynamics (dependability, structure and clarity, meaning, and impact) were built.
What Do Most People Get Wrong About Psychological Safety?
One of the most common misunderstandings is equating psychological safety with comfort or the absence of conflict. Edmondson herself has been explicit on this point: psychological safety is not about lowering performance standards or creating an environment where people are shielded from feedback. In fact, it is the opposite. Psychologically safe teams are characterized by more candor, not less; more willingness to challenge each other's ideas, to admit uncertainty, and to raise issues before they become crises.
This distinction matters enormously in practice. When leaders interpret psychological safety as "being nice" or "making everyone feel good," they tend to avoid accountability conversations, tolerate underperformance, and create cultures of false harmony. Edmondson (2019) describes a two-by-two framework in her book The Fearless Organization, plotting psychological safety against performance standards. When both are high, teams enter a "learning zone" characterized by high performance and continuous improvement. When psychological safety is high but standards are low, teams drift into a "comfort zone." And when standards are high but psychological safety is low, teams operate in an "anxiety zone," a state that produces short-term results at the cost of burnout, turnover, and suppressed innovation. When both are low, teams settle into apathy.
The anxiety zone is where many small and mid-sized businesses unwittingly find themselves. The owner or senior leader sets demanding expectations, as they should, but the interpersonal climate does not support the kind of risk-taking that learning and innovation require. People learn to keep their heads down, agree with the boss, and avoid being the one who raises a problem without a ready-made solution. Over time, this dynamic hollows out organizational intelligence. The leader makes more and more decisions in an informational vacuum, wondering why the team does not "step up."
How Does Leader Behavior Shape Psychological Safety?
Leader behavior is one of the strongest influences on whether people feel safe to speak up (Detert & Burris, 2007). In their study of leadership and team learning, Hirak, Peng, Carmeli, and Schaubroeck (2012) found that leader inclusiveness, defined as words and deeds by a leader that invite and appreciate others' contributions, was directly related to team psychological safety, which in turn predicted team learning from failures. This finding aligns with decades of I-O psychology research on transformational leadership and leader-member exchange theory, which emphasize the role of the leader in shaping interpersonal norms.
What does this look like in behavioral terms? Leaders who foster psychological safety tend to exhibit specific, observable patterns: they ask genuine questions rather than testing for "right answers," they acknowledge their own mistakes and limitations, they respond to bad news with curiosity rather than blame, and they actively solicit dissenting views before making decisions. These are not personality traits; they are learnable behaviors. But they require self-awareness, and they require feedback.
This is where multi-rater feedback becomes invaluable. A leader's self-perception of their approachability and openness is frequently disconnected from how their team actually experiences them. Research on self-other agreement in 360-degree feedback shows that many leaders rate themselves higher than their teams do, and that leaders who overrate themselves tend to be less effective (Atwater & Yammarino, 1997). Listening, inviting input, and responding well to disagreement are exactly the habits that are hard to judge from the inside. Without data from direct reports, peers, and supervisors, leaders operate on assumptions about their impact rather than evidence.
That gap is also why a quick self-check is a starting point, not a verdict. The free Psychological Safety check takes about three minutes and reads four habits a leader controls: inviting input, responding to mistakes, hearing challenge, and showing your own uncertainty. It is your own view, so use it to decide what to ask your team about, not to grade yourself.
Two FactorFactory assessments ask the people around the leader. Communication Insight measures the psychological safety a leader creates alongside how openly they share, how actively they ask for input, and how they give feedback; with its Multi-Rater Option, up to five people who work with the leader answer too, and the report sets their view beside the leader's own. The Achieving Leader 360 (AL360) looks at leadership more broadly, including how leaders handle feedback, invite participation, and respond to mistakes. For leaders who genuinely want to build psychological safety, this data is not threatening; it is the starting point.
From Practice
This composite draws on several client engagements. An environmental consulting firm with about 70 employees, a mix of scientists, project managers, and technical staff, brought in FactorFactory after struggling with a pattern that puzzled the managing partners. Individually, their senior leaders were strong. Each had a track record of delivering results and managing client relationships. But collectively, the leadership team was dysfunctional. Cross-functional projects stalled. Decisions made in leadership meetings were quietly undermined or simply ignored. Two of the strongest leaders had stopped attending optional strategy sessions altogether.
Initial conversations with the managing partners revealed the standard hypothesis: personality clashes. "They just don't get along." But the feedback told a different story. The problem was not who these leaders were. The problem was how they interacted.
When AL360 data came back, the pattern became clear. Three of the firm's most influential leaders scored noticeably lower on items related to inviting input, responding to disagreement, and creating space for dissenting perspectives. Their direct reports described environments where raising a concern was treated as a lack of commitment, where admitting uncertainty was viewed as weakness, and where the fastest way to get through a meeting was to agree and move on. One observer's comment captured it precisely: "I've learned that the safest thing to say in a leadership meeting is nothing."
The managing partners were surprised, not because they disbelieved the data, but because they had never had access to it. What followed was a structured development process that included individual coaching around specific behavioral feedback, team-level conversations about communication norms, and a follow-up assessment cycle nine months later. The firm did not transform overnight. But within a year, people outside the senior partner group were raising agenda items in leadership meetings, and the partners were hearing about problems earlier, while there was still time to act.
How Do You Build Psychological Safety With Data, Not Slogans?
The challenge with psychological safety is that it is easy to endorse in principle and difficult to build in practice. Every leader will agree that people should feel comfortable speaking up. Far fewer leaders consistently behave in ways that make speaking up feel safe. This gap between espoused values and enacted behavior is one of the oldest findings in organizational psychology (Argyris, 1991), and it is precisely why measurement matters.
Organizations serious about psychological safety need to move beyond aspiration and into assessment. This means gathering behavioral data from multiple perspectives, not just self-reports. It means tracking specific leadership behaviors, not vague competencies like "creates an open environment," but observable actions like soliciting input before stating a position, responding to errors with questions rather than blame, and publicly acknowledging when a direct report's pushback improved a decision. These are the behaviors that Edmondson's research, and the broader literature, have identified as the building blocks of psychologically safe teams.
It also means understanding that psychological safety is not a fixed state. It fluctuates with leadership transitions, organizational stress, and even meeting-to-meeting dynamics. Teams that were psychologically safe under one leader may lose that foundation entirely when leadership changes. This is why periodic assessment, not a one-time measurement, is essential. The organizations that sustain high-performing teams treat psychological safety as an ongoing practice supported by data, not a box to check during an annual engagement survey.
For small and mid-sized businesses, this is both a challenge and an opportunity. The challenge is that these organizations often lack the infrastructure (dedicated OD teams, annual climate surveys, formal coaching programs) that larger organizations use to monitor and develop team dynamics. The opportunity is that in a 50- or 150-person company, the senior leader's behavior has an outsized impact. A single leader shifting how they run meetings, respond to mistakes, and solicit feedback can change the psychological safety of an entire organization in ways that would take years in a company of 10,000.
The starting point is always the same: data. Not assumptions about how leaders are perceived, but actual multi-rater feedback from the people who experience their leadership daily. If the goal is to build teams that learn faster, adapt more readily, and perform at a higher level, the research points one way: psychological safety is a foundation. And foundations are built with evidence, not intentions.
Where to start depends on what you need. For a quick, private look at your own habits, take the free Psychological Safety check. To hear from your team, Communication Insight with the Multi-Rater Option sets their view beside yours. For a full leadership 360, use the AL360. For the everyday habits behind it, read Communication Insight: The Listening Habit That Separates Effective Leaders and Building a Culture of Psychological Safety. If you are not sure which fits, contact FactorFactory.
Sources
Research cited in this article, with links to the published work where available.
- Argyris, C. (1991). Teaching smart people how to learn. Harvard Business Review, 69(3), 99–109. https://hbr.org/1991/05/teaching-smart-people-how-to-learn
- Atwater, L. E., & Yammarino, F. J. (1997). Self-other rating agreement: A review and model. In G. R. Ferris (Ed.), Research in personnel and human resources management (Vol. 15, pp. 121–174). JAI Press.
- Detert, J. R., & Burris, E. R. (2007). Leadership behavior and employee voice: Is the door really open? Academy of Management Journal, 50(4), 869–884. https://doi.org/10.5465/amj.2007.26279183
- Duhigg, C. (2016, February 25). What Google learned from its quest to build the perfect team. The New York Times Magazine. https://www.nytimes.com/2016/02/28/magazine/what-google-learned-from-its-quest-to-build-the-perfect-team.html
- Edmondson, A. C. (1996). Learning from mistakes is easier said than done: Group and organizational influences on the detection and correction of human error. The Journal of Applied Behavioral Science, 32(1), 5–28. https://doi.org/10.1177/0021886396321001
- Edmondson, A. C. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383. https://doi.org/10.2307/2666999
- Edmondson, A. C. (2019). The fearless organization: Creating psychological safety in the workplace for learning, innovation, and growth. Wiley.
- Frazier, M. L., Fainshmidt, S., Klinger, R. L., Pezeshkan, A., & Vracheva, V. (2017). Psychological safety: A meta-analytic review and extension. Personnel Psychology, 70(1), 113–165. https://doi.org/10.1111/peps.12183
- Hirak, R., Peng, A. C., Carmeli, A., & Schaubroeck, J. M. (2012). Linking leader inclusiveness to work unit performance: The importance of psychological safety and learning from failures. The Leadership Quarterly, 23(1), 107–117. https://doi.org/10.1016/j.leaqua.2011.11.009
