Lean works. It is documented in decades of research and on factory floors around the world. And yet we hear it again and again: "We tried Lean — it didn't work for us." That is a misdiagnosis, and it is expensive — not because it costs a consulting project, but because it leaves the organisation with the wrong lesson.
What actually runs aground
Lean programmes don't run aground on 5S, VSM or board meetings. They run aground on what is required to keep them alive: management's daily behaviour. When top management approves a Lean initiative and then carries on with the same meeting structures, the same measurement systems and the same reactive logic — they send a signal far stronger than any kick-off meeting. The signal is: this is something for production, not for us.
MIT researchers Nelson Repenning and John Sterman have described the mechanism precisely in what they call the capability trap: under pressure, organisations revert to what they know — quick fixes and heroics — rather than investing in the systems and routines that, over time, make problems rarer. The paradox is that the more there is burning, the harder it is to spend time building the fire alarm. And the more rarely you build the fire alarm, the more things keep burning.
"Under pressure, organisations revert to what they know — and that prevents them from building what removes the pressure."
— Repenning & Sterman, MIT
The visible and the invisible leader
There are two kinds of leader in a Lean effort. The visible leader is the one at the kick-off, approving the budget and attending the status meeting. The invisible leader is the one who decides what actually happens when no one is looking: who gets praised, what gets deprioritised, which problems get commented on and which simply slide past.
Lean requires the invisible leader. The one who actually shows up at the board. Who asks questions instead of dictating answers. Who treats a deviation note as information rather than criticism. The Lean tools are only aids. It is the management behaviour that decides whether they hold.
Deming's point that never went out of fashion
W. Edwards Deming argued systematically that the vast majority of problems in an organisation are not caused by the employees — they are caused by the system management has built and is responsible for. His conclusion: you can't improve an organisation by talking to the employees. You improve it by changing the systems and structures that shape their behaviour. And that is always management's job. Shook's NUMMI example illustrates the point: the same employees who previously produced GM's worst cars produced, with Toyota's management system, some of the best. The people were not the problem.
What it takes in practice
This is not an attack on managers. It is a clarification of what the task actually is. A Lean programme is, in reality, a leadership development programme. It requires managers to change their own routines — not just the employees'. To be present at the board meetings, not because they are invited, but because they have made it part of what it means to lead in this organisation.
When a Lean programme doesn't hold, the question is never: "Did Lean work?" The question is: "What happened to the management behaviour when the project ended?" The answer reveals everything.
✓The most important takeaways
- →Lean doesn't fail. Lean programmes fail because the management behaviour doesn't change enough.
- →The strongest signal management sends is what they do — not what they say.
- →The capability trap: under pressure we revert to what we know — and that prevents us from building what reduces the pressure.
- →A Lean effort is, in reality, a leadership development programme. Treated as a production project, it will run aground.
Sources
- Repenning, N. P. & Sterman, J. D. (2001). Nobody Ever Gets Credit for Fixing Problems That Never Happened. California Management Review, 43(4), 64–88.
- Deming, W. E. (1986). Out of the Crisis. MIT Press.
- Shook, J. (2010). How to Change a Culture: Lessons from NUMMI. MIT Sloan Management Review, 51(2).
The references show the professional basis. Sources are reproduced in their original language.