The best-performing hospital teams in Amy Edmondson’s landmark research reported more medical errors than the worst-performing ones. That finding, from her 1999 study of nursing units, upended a common assumption. High-performing teams weren’t making more mistakes. They were surfacing them. The weaker teams were burying them, and paying the price in patient outcomes nobody dared explain.
That paradox is the diagnostic problem every HR leader now faces. A workplace where nobody admits errors, disputes a manager’s plan, or raises an early warning isn’t necessarily well-run. It may just be quiet. The interesting question isn’t whether people at a company make mistakes. They do, everywhere. The question is what happens next, and who inside the organisation can tell.
What “Safe to Learn From Mistakes” Actually Means
Psychological safety, as Edmondson defined it in her Administrative Science Quarterly paper, is a shared team belief that interpersonal risk-taking will not be punished with humiliation, ridicule, or career damage.
It isn’t comfort. It isn’t the absence of accountability. It’s the specific condition under which admitting an error, questioning a decision, or asking a naive question feels professionally survivable. The foundational psychological safety concept has moved from academic construct to boardroom concern because the operating cost of its absence is now measurable.
Google’s Project Aristotle, a multi-year study of team effectiveness inside the company, ranked psychological safety as the single strongest predictor of team performance, above dependability, structure, or seniority. In plain terms: teams that can safely say “I got that wrong” outperform teams that can’t, on the same problem set, with the same talent.
Signals That Reveal What Kind of Culture It Really Is
A learning-tolerant workplace and a blame-driven one don’t announce themselves in policy documents. They show up in small, repeatable behaviours, most of them visible inside the first thirty minutes of any project review. The signals below show up as behavioural contrasts, and every HR team already has access to them without commissioning a survey.
| Behavioural signal | Blame-tolerant workplace | Learning-tolerant workplace |
| A junior engineer flags a design flaw | Silence in the room; the concern is escalated privately later | The question is treated as data; the team pauses |
| A quarterly target is missed | Deflection, explanations, restructuring of the metric | A written post-mortem is circulated within a week |
| A senior manager’s plan looks weak | Nobody says so in the meeting | At least one report disagrees on record |
| An intern breaks a production system | Fired or informally exiled | Root-cause review; the process gap is owned upward |
| The word “mistake” appears in a review | Only in performance appraisals | In weekly team meetings, without drama |
None of these signals is complicated. What makes them diagnostic is that they aggregate. A team where several of these signals tilt toward the left column is not a team with a training problem. It’s a team where speaking honestly costs more than staying silent, and where silence has been rationally chosen.
Why Indian Workplaces Have a Sharper Version of This Problem
Indian workplaces intensify the same diagnostic gap through familiar structural pressures. Hierarchical distance between senior leaders and frontline staff tends to be wider than in flatter organisational cultures, and the personal cost of visible failure travels further.
When Infosys terminated roughly 800 trainees across four rounds in 2025 for failing internal assessments, the message reaching every other trainee across the industry wasn’t only about assessment standards. It was about what happens to those who don’t clear the first bar.
The mental-health data supports the pattern. Deloitte India’s Mental Health in the Workspace 2022 survey, covering 3,995 employees across twelve industries, found that 47% of respondents identified workplace stress as the largest single factor affecting their mental well-being, with the associated productivity and attrition cost estimated at $14 billion a year. A separate strand of the same problem shows up in India’s still-limited whistleblowing culture, where the structural gaps in speak-up protection keep even legitimate concerns off the record.
Both patterns compound. Managers pattern-matched to hierarchy, and employees calculating the personal risk of speaking up inside that hierarchy, produce the subtle behaviours that quietly suppress team learning even where the company’s stated values say the opposite.
The Honest Audit
An honest audit doesn’t begin with an engagement survey, because the same fear that suppresses candour in meetings suppresses it on the form. It begins with artefacts already sitting inside the organisation’s own systems, most of them accessible without commissioning any new work. The metrics below can be pulled in a working week by anyone with access to the project-management tool, the incident database, the exit-interview archive, and the manager’s calendar.
Post-mortem volume and circulation
The number of documented post-mortems written in the last quarter, cross-referenced with how many were read or annotated outside the team that wrote them. A high write-rate paired with low external readership indicates a team documenting for compliance rather than for organisational learning.
Repeat-incident rate
The share of incidents in the current quarter that duplicate the root cause of an incident from the previous four quarters. A rising rate is the clearest evidence that post-mortem findings aren’t being absorbed, which usually means the process treats them as ceremony rather than as inputs into the next planning cycle.
Manager-override frequency
The count of decisions reversed by data or dissent raised inside a team review versus in a private 1:1 afterwards. Reversals surfacing in the open are a positive signal. Ones happening only in 1:1s point to a hierarchy where public disagreement is quietly punished, even when the stated culture claims otherwise.
Attrition-and-error-reporting overlay
The list of teams sitting in the top decile of error reporting alongside the bottom decile of voluntary attrition. That intersection is where a healthy learning culture actually lives inside the company, and the practices of those teams are transferable to the rest of the organisation without a consultant.
Exit-interview theme cluster
The frequency of phrases like “wasn’t heard”, “raised it and nothing happened”, or “no one wanted to know” across the last twelve months of exit interviews. The early signals of a broken safety climate usually appear here years before the culture survey catches them, and the people who leave first are often the ones who tried to speak up first.
In the End…
Workplaces that survive the next decade of AI-driven change, faster product cycles, and thinner margins won’t be the ones with the cleanest incident reports. They’ll be the ones where the incident reports get written at all. HR leaders who want an early reading on their own organisation can begin with a week of quiet observation.
Sitting in on three project reviews and three post-mortems, and counting how many times a junior person contradicts a senior one without preamble, will surface more than most engagement dashboards. If the count is zero, the diagnostic is already done, and the next step is rebuilding the feedback loop that made the silence rational in the first place.
FAQs
What is psychological safety at work?
Psychological safety is a shared team belief that speaking up with an error, a question, or a dissenting view will not lead to humiliation or career damage. Amy Edmondson introduced the term in her 1999 study of hospital nursing units and defined it as the specific condition under which interpersonal risk-taking feels professionally survivable.
How can HR leaders measure psychological safety without running a survey?
The most reliable read comes from artefacts already inside the company: the volume of documented post-mortems, the share of incidents in the current quarter that repeat a prior root cause, how often manager decisions are reversed in a review versus in a 1:1, and recurring exit-interview phrases such as “wasn’t heard” or “raised it and nothing happened.”
Why is psychological safety harder to build in Indian workplaces?
Hierarchical distance between senior leaders and frontline staff is wider than in flatter cultures, and the visible cost of failure travels further. Deloitte India’s 2022 survey of 3,995 employees found 47% flagged workplace stress as the single largest factor affecting mental well-being, with the productivity and attrition cost pegged at $14 billion a year.
What is the difference between a blame-tolerant workplace and a learning-tolerant one?
A blame-tolerant workplace goes silent when a junior contradicts a senior, deflects on missed targets, and exits interns who break production. A learning-tolerant workplace treats the same events as data, circulates a written post-mortem within a week, and owns the process gap upward.
Does psychological safety mean lower accountability?
No. Psychological safety removes the interpersonal penalty for admitting an error, so the error surfaces early enough for the team to act on it. Accountability still applies to the standard of work, the decision made, and the response to what surfaced.
