Why Leadership Training Fails (And What To Do Instead)

"Your leaders already know what good leadership looks like. The gap isn't knowledge. It's what they do at 4pm on a Thursday when everything is on fire."

The awkward part is that it does work. Leadership training works brilliantly... for the people who were already going to change. Everyone else goes back to their desk exactly as they were.

You know the shape of it. The workshop lands well. The scores are lovely. Three months later the director who dominates every meeting is still dominating every meeting. The manager who ducks conflict is still ducking conflict. The one who decides everything alone is still deciding everything alone. Nobody is being dishonest. Nothing simply happened.

And it is not a content problem. The frameworks are usually sound. The facilitators are usually good. The enthusiasm on the way out is usually real. The problem sits underneath all of that, in the design, and until L&D sees it we will keep building excellent programmes that change nothing.

The Knowing-Doing Gap

Jeffrey Pfeffer and Robert Sutton wrote a whole book about this in 2000. The Knowing-Doing Gap. Leadership development has been politely ignoring it ever since.

Their point was simple. Knowing what to do and doing it are two completely different acts, and the link between them is much weaker than any organisation wants to believe.

Try it. Ask a room of senior managers whether a leader should listen more than they talk in a one-on-one. Every hand goes up. Now ask them to walk you through a conversation from last week they felt good about, and count how much listening actually happened in it. Watch the room get quiet.

That gap isn't lying. It is what happens when the calendar is full, the inbox is screaming and something needs deciding now. Under pressure people don't do what they know. They do what they can already do without thinking.

Most training teaches the aspiration and then sends everyone home. The behaviour never gets a chance to form.

What Training Is Actually Good At

Three things, and it does them well. Awareness. Vocabulary. Intention. People leave understanding the idea, able to name it, and genuinely meaning to use it.

That is not nothing. A team that can say "we're doing the thing again" mid-argument is ahead of a team that can only feel that something is off.

But awareness is not behaviour. Behaviour is what you do when the budget is short, the brief is vague, the person across the table wants something else, and the call has to be made before anybody has the full picture. Behaviour lives at the moment of pressure.

And pressure is exactly the thing training strips out. Deliberately. To keep the room comfortable.

A workshop has no pressure in it. A case study has none. A role-play has the ghost of it, just enough that everyone knows they are performing rather than actually caught in something. So the learning stays where the conditions are, which is the training room. Put people back in the real building with real stakes and the old behaviour is waiting for them at their desk.

The Three Things That Actually Move Behaviour

The research is boringly consistent on this. Consequence. Feedback. Repetition in conditions that look like the real thing.

Consequence. A choice has to produce something you can see and feel. In a workshop, a manager makes a call and... nothing. The facilitator nods. The slide advances. In a world that reacts, a bad call comes back and finds you. That leaves a mark. Nodding does not.

Feedback. You have to be able to see what you actually did, not what you meant to do. Everyone has a blind spot here. People who believe they are collaborative and are quietly territorial. People who believe they are decisive and are just impatient. Discussion won't find that. Only watching yourself do it will.

Repetition in context. Practise a behaviour only in a training room and it stays attached to the training room. It doesn't follow anyone back to the office. It has no reason to.

Ordinary workshops remove all three, and I do understand why. A room where people can fail properly is a hard room to sell. So we make it safe, and the safety is what takes the pressure out.

The Measurement Makes It Worse

Ask people if they enjoyed it. Ask if the facilitator was good. Ask if they felt they learned something. All fine questions. All measuring the quality of an afternoon, not the durability of a behaviour.

And the survey usually goes out within a day or two, while the intention is still warm and nobody has been tested yet. Of course the numbers are high. The numbers were always going to be high.

So the loop closes on itself. Good scores get read as proof it worked. The programme runs again. Nothing shifts at scale. And when someone finally asks why, the answer is "poor implementation" rather than "we designed a thing that could never have worked."

Leadership Is Not a Solo Sport

Behaviour isn't only in the person. It is in the room around them. The manager avoids conflict partly because of who they are and partly because conflict on that team has always turned nasty. The director hoards decisions partly by temperament and partly because delegating went badly once and everyone has since adapted to expect the hoarding.

Train that person alone, or in a room of strangers from other companies, and you can surface the pattern but never the system holding it in place. Which is why anything involving real teams, generating real social friction, beats the alternative almost every time.

So What Actually Works

Not abandoning leadership development. Designing it so the three conditions are actually present.

Serious games are one of the cleanest ways to get there without putting anybody's real quarter at risk. Decisions produce visible consequences. Behaviour is out in the open where the facilitator and the whole team can see it. And the team's real dynamics show up inside two hours, because a game is a pressure cooker with rules. Then the debrief connects what just happened to what happens every Monday.

It is not the only route. Good executive coaching changes behaviour. A stretch assignment with real stakes changes behaviour. An action learning set chewing on a live problem changes behaviour. What they all have in common is consequence, feedback, and something that resembles the real environment.

The organisations that get the most out of this are the ones honest about the limits. Training raises awareness. It builds vocabulary. It gives people a shared way to talk. What it cannot do on its own is change what a leader does in the meeting, in the hard conversation, on the call made with half the information.

A Better Question to Bring to the Table

Most L&D teams ask: what is the best leadership programme we can run?

The better question: which specific behaviours do we need to change, and what conditions would actually produce that change, here, with these people?

Start at the behaviour and work backwards to the design. You will end up somewhere completely different from the catalogue. You will also have some harder conversations with stakeholders who assumed good training reliably produces good outcomes. Worth it. Results are the whole reason anyone signed the cheque.

Leadership training fails when it is built for the training room instead of the building people actually work in. The fix is not a better programme. It is a different idea of what changes a person, and a design built on that instead.

Want to see what that looks like on a Tuesday with your team in the room? Write to me. Or have a look at the games that make those conditions happen.

With love from Bengaluru, this is Arvindh saying over and out.