Psychological safety has become a buzzword, posted on LinkedIn, printed on culture decks, invoked in exit interviews. Most organisations that claim it do not have it. Amy Edmondson's definition is precise: "a shared belief that the team is safe for interpersonal risk-taking." Not "nice." Not "comfortable." Risk-taking. Speaking up with a dissenting view. Admitting a mistake. Asking a naive question. Challenging a senior person. If those actions carry social or career cost, there is no psychological safety. There is only politeness.

The four stages, not a binary

Timothy Clark's framework is useful: Inclusion safety (I belong). Learner safety (I can ask). Contributor safety (I can offer). Challenger safety (I can disagree). Most Indian teams stall at Inclusion, "we are family", and mistake it for the whole thing. Challenger safety is where innovation, and error prevention live. it is also the hardest because it requires the most powerful person in the room to model it first.

What it looks like in a meeting

A junior engineer says "I think this approach has a scalability issue." The tech lead says "Talk me through it", not "We've already decided" or "Do you know how long I've been doing this?" The team watches. If the lead listens, the norm strengthens. If the lead dismisses, the norm dies. That moment, repeated across hundreds of interactions, is the culture. No poster changes it. Only behaviour does.

The Indian context: hierarchy is the default

Respect for seniors is cultural infrastructure. it is not inherently toxic. But when "respect" means "do not question," psychological safety cannot exist. The organisations that build it in India do so explicitly: "We value hierarchy for clarity, and we value dissent for quality. Both can coexist." They create structured dissent mechanisms, pre-mortems, red teams, anonymous input channels, "disagree, and commit" norms. They do not hope for it. They engineer it.

Measurement that is not a survey

Edmondson's 7-item scale is the standard. But the real signal is behavioural: How many ideas from junior members get implemented? How many near-misses are reported vs. hidden? How long does it take for a bad decision to be challenged? WHO speaks in retrospectives, the same three voices or the whole team? These are observable. Track them.

What leaders do to build it

Admit their own mistakes publicly. "I was wrong about the timeline. Here is what I missed." Ask "What am I missing?" and wait. Protect the dissenter when the room turns. Reward the bad news bearer, "Thank you for catching this early." Model not-knowing, "I don't have the answer. Let's figure it out." When I was consulting inside a large corporate EAP program a few years ago, the leaders WHO did these things had teams that surfaced problems in weeks. The ones WHO didn't heard about them in exit interviews.

Building challenger safety in high power-distance cultures

In Indian teams where the senior person speaks first and last, challenger safety requires engineered structures. Pre-mortems work because they legitimise dissent: "We have failed. Why?" is a licensed question. Red teams work because the role is assigned, you are paid to disagree. Anonymous input channels work because they remove the social cost. "Disagree, and commit" works because it separates voice from veto.

The leader's behaviour is the only thing that scales. If the VP says "I was wrong" in the all-hands, the manager says it in the team meeting, the tech lead says it in the stand-up. If the VP shoots the messenger, no pre-mortem survives. I have seen a CEO in Bangalore institute a "failure of the month" award, not for the mistake, for the learning shared. The cultural shift took two years. It started with the CEO's first story.

What destroys it instantly

Shooting the messenger. Retaliation disguised as "performance management." Meetings where the decision is pre-made and the discussion is theatre. Leaders WHO confuse psychological safety with consensus, "we all agree" is not the goal. "We all got heard" is.

Building challenger safety in high power-distance cultures

In Indian organisations, hierarchy is the default. The junior engineer does not contradict the senior architect. The team member does not challenge the manager's timeline. Challenger safety (the permission to say "this will not work") must be explicitly built. The practices that work: the manager states "I expect you to disagree with me" in the first team meeting. The manager models vulnerability: "Here is a recent decision I got wrong." The retrospective format requires "one thing I would change about my own proposal." The all-hands has a "dissent slot" where anyone can raise a concern anonymously. When a Bangalore product team implemented these, their sprint retrospective dissent rate went from 0 to 4.2 per session in 8 weeks. The project delivery timeline improved 15%. Safety is not comfort. it is the permission to be uncomfortable together.

What destroys it instantly

One public reprimand. One "I told you so" after a failed experiment. One manager WHO says "bring me solutions, not problems" and then punishes the solution that differed from their own. One leader WHO quotes psychological safety in the town hall and then ignores the dissent in the review meeting. The gap between the spoken value and the lived behaviour is where trust evaporates. It takes 50 consistent interactions to build safety. It takes one inconsistent interaction to destroy it. The manager WHO understands this protects the team's safety like a budget line item. The manager WHO does not loses the team's best thinking first, then the team.

The Indian context: hierarchy as the default, not the exception

Western psychological safety research assumes flat teams. Indian teams assume hierarchy. The adaptation is not "flatten the hierarchy." The adaptation is "make hierarchy safe." The senior architect WHO says "I need your eyes on this design because I might miss something" creates more safety than the manager WHO says "we are all equals here" but acts otherwise. The cultural lever is respect, not equality. The framework that works: respect the role, challenge the assumption. The junior engineer challenges the technical assumption. The senior architect respects the challenge. The hierarchy holds. The safety grows.

The EAP consultant's view: surveys lie, behaviour tells the truth

When I was consulting inside a large corporate EAP program a few years ago, the quarterly engagement survey showed "psychological safety: 3.8/5." The same team had zero dissent in sprint retrospectives, zero questions in all-hands, and 40% voluntary attrition of high performers in 18 months. The survey was wrong. The behaviour was right. The metrics that actually track psychological safety: dissent rate in retrospectives (target >2 per session), question rate in all-hands (target >5 per session), time-to-first-disagreement in new teams (target <2 weeks), voluntary attrition of high performers (target <10% annually), and internal innovation proposals per quarter (target >3 per team). These are observable. They do not require a survey. They require a leader WHO watches.

If this sounds familiar and you want to work through it properly rather than just read about it, that's exactly the kind of thing I work on with clients. Book a Session For EAP fundamentals, see What Resilience Training Actually Looks Like Inside a Company and for programme design mistakes, see Why High-Performing Teams Still Burn Out: Lessons From Corporate Consulting.

Written by Vishal Ram, Consultant Psychologist and Founder of WayToTherapy. Vishal is an Affiliate & Consultant Psychologist with Lyra Health, a member of the MantraCare provider network, and an International Affiliate Member of the American Psychological Association. He has worked across corporate mental-health consulting (including Accenture's Project META), group facilitation, and private practice in Mumbai. Book a session or read more about Vishal.