TL;DR: Lean doesn't fail because people resist change. It struggles when organizations ask for participation without earning belief. Don Ephlin saw this decades ago: behavior changes when people believe, not when they're told to obey. That distinction still explains why some Lean efforts transform cultures while others quietly settle into compliance.
I've been working my way through the Don Ephlin papers for a while now. Ephlin, the late UAW vice president and, for what it's worth, a professor of mine at MIT, was thinking hard about trust, respect, and labor-management partnership long before “Lean” became a label we put on things. The longer I spend with his writing, the more I'm convinced he understood something that continues to trip up organizations today.
So this is another post in what I've been calling the Ephlin Files. And it's about a pattern I keep running into.
Activity Isn't the Same as Transformation
You've probably seen the version of this I'm describing.
An organization invests in Lean. People get trained. Kaizen events run. Visual boards go up. Metrics get tracked. From the outside, it looks like progress.
And then… it doesn't quite stick. The activity is there. Results are uneven. The culture doesn't really change.
People are doing what's being asked of them. They show up to the huddle. They fill out A3s. They participate in improvement events. But participation isn't the same as engagement, and engagement isn't the same as belief.
Sometimes what looks like adoption is actually compliance.
An Ephlin Line That Doesn't Let Go
There's a line I keep returning to:
Behavior changes when people believe, not when they're told to obey.
It sounds almost obvious. But sit with it for a minute and a harder question comes out of it:
Are we building systems that create belief, or systems that depend on compliance?
Because those produce very different organizations.
What Compliance Looks Like
Compliance can be hard to see from across the room. The boards are updated on schedule. The meetings happen. The right words get said.
Underneath, there's hesitation. People share the safe problems, not the real ones. Ideas get filtered before they're spoken out loud. Concerns about workload, safety, or job impact go unsaid. Not because people don't care, but because they're not sure what happens if they do speak up.
That isn't resistance. It's a rational response to what the system has taught them.
What Shapes the System
It's easy to ask, “how do we get more engagement?” But engagement isn't something you can order up. It gets shaped by experience.
What happened the last time someone raised a real problem? How do leaders respond when the numbers are missed? Are mistakes treated as something to learn from, or something to explain away?
People are reading these signals constantly. Over time, the signals either build belief or erode it.
NUMMI: What Actually Made It Work
When people talk about NUMMI, the conversation usually starts with the tools. Standard work. Andon. Team-based problem solving. That's the surface.
The deeper story, as I've written about before, is about trust and relationships. Workers had reason to believe that speaking up wouldn't get them in trouble. That improving the process wouldn't eliminate their jobs. That leadership was willing to listen and not just direct.
That didn't come from a training program. It came from consistent leadership and union behavior over time. It wasn't perfect. But it was different enough to matter, and GM's own 1987 internal report on NUMMI said so plainly.
When Belief Is Missing
When Lean is experienced as something people have to do, a few things tend to follow.
Problems get managed quietly instead of surfaced openly. “Improvement” becomes about meeting expectations rather than actually improving anything. Energy shifts from learning to self-protection. And continuous improvement starts to feel like continuous pressure.
That's a tough foundation for anything sustainable.
The Role of Stability
One piece that doesn't get enough weight in Lean conversations is stability.
If people believe that improvement might cost them their job, or expose them to unfair treatment, that changes how they participate. Or whether they participate at all.
This is where labor-management relationships matter more than most Lean discussions acknowledge. Structures that protect fairness and give people a real voice don't slow improvement down. They make it possible. Ephlin understood that. And the Ford/UAW delegation that went to Japan in 1981 came home saying much the same thing, though not everyone back in Detroit wanted to hear it.
Better Questions to Ask
If Lean isn't taking root, the instinct is to ask why people aren't embracing it. A more useful question might be: what are people experiencing that makes belief difficult?
Are we asking for input but reacting defensively when we get it? Are we promoting improvement while creating fear about its consequences? Are we going to the gemba but not actually listening when we're there?
These aren't comfortable questions. They're more honest ones.
A Closing Thought
Maybe Lean doesn't fail because people resist change. Maybe it struggles because belief hasn't been earned.
Belief doesn't get built through communication plans or training sessions. It gets built through everyday actions, in how we respond to problems, how we treat people when things go wrong, how consistently our behavior matches our stated values.
People notice. They always do.
If Ephlin was right that behavior changes when people believe, then the question worth sitting with isn't whether the tools are being used correctly. It's this:
What are we doing, intentionally or not, that makes belief possible, or impossible?
That's a harder place to start. It's probably also where real improvement begins.
This post is part of an ongoing series drawing on the Don Ephlin papers. You can find the rest of the Ephlin Files here.
