It’s a Performance Metric
Here is a thing that happens in organizations with surprising regularity: a leader invests in a psychological safety initiative, it gets labelled a ‘culture program,’ and it gets quietly defunded when the next budget cycle comes around. Meanwhile, the same organization is haemorrhaging talent, sitting on problems nobody will name out loud, and wondering why its most capable people seem to be operating at about sixty per cent.
These things are connected. And the research has been saying so, with increasing specificity, for more than twenty-five years.
Psychological safety – the shared belief that a team is safe for interpersonal risk-taking, that you can ask a question, admit a mistake, or challenge an idea without fear of embarrassment or punishment – is not a nice thing to have when the business is doing well. It is a structural condition for performance. And in 2026, in workplaces that are more complex, more distributed, and more dependent on human judgment and adaptability than ever before, it matters more than ever.
What the research actually shows
Amy Edmondson coined the term ‘team psychological safety’ in her 1999 paper in the Administrative Science Quarterly and has spent the decades since methodically demonstrating that it is a performance variable, not a comfort variable. Her 2023 review with Bransby in the Annual Review of Organizational Psychology synthesized the now substantial literature: psychological safety enables learning behaviour, knowledge-sharing, error-reporting, and the kind of constructive challenge that prevents bad decisions from becoming expensive ones.
The mechanism is straightforward. When people in an organization believe that speaking up will be penalized – even subtly, even unintentionally – they stop speaking up. Information stops flowing. Problems get hidden until they become crises. Ideas get shelved before they are ever said aloud. The organization becomes, in a very practical sense, dumber than the people in it.
Google’s Project Aristotle made this concrete in a way that reached leaders who had not been reading the I/O psychology journals. Between 2012 and 2014, the company studied 180 of its own teams across 200 interviews and 250 team attributes, trying to identify what made teams effective. The answer, at the top of the list, was not technical skill, tenure, or team composition. It was whether team members felt safe enough to speak up.
A team of A-players operating in a low-safety environment will consistently underperform a team of B-players who trust each other enough to be honest. Talent is necessary. It is not sufficient.
The cost of getting this wrong is not abstract. Gallup’s 2026 State of the Global Workplace report found that only 20 percent of employees worldwide were engaged in 2025 – and that disengagement is costing the global economy an estimated $10 trillion in lost productivity. Engagement and psychological safety are not identical, but they are tightly coupled. You cannot build or sustain genuine engagement in an environment where people are managing their self-presentation instead of doing their work.
The neurodivergent angle that most leaders are missing
The standard psychological safety conversation tends to focus on what happens when anyone feels unable to speak up. That is accurate and important. But there is a specific, largely underdiscussed cost that falls on employees with ADHD and other neurodivergent employees in low-safety environments – one that compounds the general productivity loss considerably.
It is called masking, and it is exactly what it sounds like: the active, effortful suppression of natural ways of thinking, communicating, processing, and working to appear neurotypical. People with ADHD and autistic employees often learn to mask early, because the alternative – being seen as difficult, distracting, or ‘not a fit’ – carries real professional consequences.
In a low-safety workplace, masking becomes more intense. Every interaction becomes a calculation. Every contribution is filtered through the question: ‘Is it safe to say this the way I would naturally say it?’ That calculation consumes working memory and cognitive bandwidth that would otherwise go toward the actual work. The I/O psychology research on psychological safety and threat responses at work is directly relevant here. When the brain perceives social threat, such as the threat of judgment, exclusion, or negative evaluation, it activates the same threat-response circuitry as physical danger. Cognitive resources narrow. The prefrontal cortex, responsible for flexible thinking, creative problem-solving, and executive function – key elements that knowledge work depends on – becomes less available. For neurotypical employees, this is a cost. For employees with ADHD, who are already managing executive function challenges, it is a tax on a resource that was already stretched
The creative, divergent, hyper-focused problem-solving that makes ADHD thinkers genuinely valuable to organizations never fully shows up in a low-safety environment. That is not a personal failing. It is a design failure.
Psychological safety, in this context, is not just a nice condition for neurodivergent employees. It is the structural condition that allows them to do the work you hired them to do. Organizations that invest in it are not being generous. They are being sensible.
A necessary clarification
Every time psychological safety comes up in a leadership conversation, someone in the room has the same thought: ‘But does this mean we can’t hold people accountable?’
It does not. And Edmondson has been making this point for decades, because the misunderstanding is so persistent.
Psychological safety is not the absence of standards. It is not a permission slip to miss deadlines, avoid hard feedback, or let underperformance slide. High-safety teams do not have lower standards. They have higher ones, and they are more likely to meet them, because people can flag problems early, ask for help without shame, and challenge a bad idea before it becomes an expensive mistake.
The confusion arises because ‘safety’ and ‘comfort’ sound similar. They are not. Psychological safety is about interpersonal risk, not task difficulty. A team where everyone trusts each other enough to say ‘I think we’re getting this wrong’ is not a comfortable team. It is a high-functioning one.
Without safety, high standards become a source of fear. And fear is a terrible long-term management strategy. (It produces short-term compliance and long-term disengagement, which is precisely the pattern Gallup has been measuring.)
Four things leaders can actually do
The research is clear on what builds psychological safety. None of it is complicated. All of it requires intention.
The question worth sitting with
Here is a useful diagnostic if you lead a team:
When did someone last tell you something you did not want to hear?
If you can recall a specific instance recently – good. If you are searching and coming up blank, that is not evidence your team has no bad news. It is evidence your team does not believe it is safe to bring it to you. And somewhere in that gap is a problem no one is solving, a mistake no one is correcting, and a perspective no one is offering – because the environment made withholding the safer choice.
Psychological Safety is not the whole of inclusive leadership, but it is fundamental to it. Inclusive leadership is not just a social or moral posture; it is an operational one. In for-profit organizations, it makes sense for the same reason any sound management practice does: it improves the quality of thinking, increases the likelihood that critical information will surface in time, and allows more of your people to contribute at full capacity. Without that foundation, organizations do not just become less equitable. They become less effective, less productive, and less able to translate the diversity they have hired into actual business performance. The good news is that it is buildable. Not through a program, not through a one-day workshop, but through a consistent, daily practice of leadership behaviour that signals, again and again: it is safe to tell the truth here.
That signal is free to send. And the cost of not sending it is, as Gallup recently calculated, roughly $10 trillion a year. It seems like a reasonable place to start.
Does this article strike a chord with you? You can download a PDF version that can easily be shared with HR practitioners, people managers and senior leaders – we all benefit when we can show up as our best selves.
Sources
- 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., & Bransby, D. P. (2023). Psychological safety comes of age: Observed themes in an established literature. Annual Review of Organizational Psychology and Organizational Behavior, 10, 55–78. https://doi.org/10.1146/annurev-orgpsych-120920-055217
- Google. (2016). Project Aristotle: Understanding team effectiveness. Re:Work. https://rework.withgoogle.com/intl/en/guides/understand-team-effectiveness
- Gallup. (2026). State of the global workplace: 2026 report. Gallup Press. https://www.gallup.com/workplace/349484/state-of-the-global-workplace.aspx
- Capezio, A., Barends, E., Rousseau, D., & Wietrak, E. (2024). Psychological safety: An evidence review. Chartered Institute of Personnel and Development (CIPD). https://www.cipd.org/globalassets/media/knowledge/knowledge-hub/evidence-reviews/2024-pdfs/8550-psychological-safety-scientific-summary.pdf