KOHLS CONSULTING
Menu
NUMMI’s Hidden Lesson: Why Habits, Not Culture, Drove Transformation
When people talk about the famous NUMMI experiment — GM and Toyota’s joint venture in Fremont, California — the story usually goes like this:
GM’s worst plant became its best. Absenteeism plummeted, quality soared, and productivity jumped. It was all about “culture” and “mindset change.”
But here’s what almost nobody says: NUMMI didn’t start with mindset. It started with habits.
Two Loops, Two Worlds
Every habit follows a simple loop: Trigger → Action → Reward.
At GM Fremont before NUMMI, the loop was cortisol-driven:
It was toxic, but predictable. Workers could survive in that loop. And survival was enough.
Toyota brought a different loop. At NUMMI, workers agreed to UAW pay — but only if they worked Toyota’s way:
The Gate Called Predictive Adaptive Capacity
In my book Habit Engineering, I call this decision point Predictive Adaptive Capacity (PAC):
“Is it safe to try something new? Will the reward outweigh the risk?”
At GM, the answer was no. Pulling the cord wasn’t safe — it invited punishment. So PAC was low, and the learning loop never opened.
At NUMMI, the answer was yes. Pull the cord, and you got support. PAC was high, so the learning loop activated. And once learning starts, behavior changes.
Why Japan Mattered
Habits don’t stick in a week. They take six months to two years to really embed. That’s why Toyota did something radical: they sent Fremont workers to Japan for a year.
No lectures. No posters. Just the line. Their hands learned first. They pulled cords, solved problems, and were rewarded with help instead of blame. Slowly, the old muscle memory gave way to new habits.
When they returned to Fremont, those habits had weight. And when reinforced inside a Toyota-run environment, they stuck.
Why GM Couldn’t Copy It
Here’s the overlooked truth:
So while Toyota transferred habits back home, GM only took notes. And as one Toyota executive later said: “We showed them everything, and they still didn’t change.”
The reason was simple: GM didn’t need more knowledge. They needed new habits. And they had no safe place to build them.
What This Means for YouNUMMI wasn’t about copying tools. It wasn’t about waiting 30 years for culture to “emerge.” It was about engineering safe, rewarding habit loops long enough for them to stick.
If you want TPS-level transformation, the lessons are clear:
Conclusion
NUMMI succeeded because Toyota didn’t start with mindset. They dismantled cortisol-driven loops and replaced them with dopamine-driven ones in a safe environment — long enough for new habits to stick.
That’s the hidden lesson most companies miss. And it’s the reason so many change efforts fail.
If you’ve ever wondered why your own Lean or TPS effort stalled — and what it takes to make change stick — that’s exactly what I explore in my book, Habit Engineering.
Because the real question isn’t whether your people “believe” in change. The question is: have you engineered the loops that make belief inevitable?
GM’s worst plant became its best. Absenteeism plummeted, quality soared, and productivity jumped. It was all about “culture” and “mindset change.”
But here’s what almost nobody says: NUMMI didn’t start with mindset. It started with habits.
Two Loops, Two Worlds
Every habit follows a simple loop: Trigger → Action → Reward.
At GM Fremont before NUMMI, the loop was cortisol-driven:
- Trigger: A broken process or another top-down mandate.
- Action: Cut corners, protect yourself, avoid blame.
- Reward: Relief, and a paycheck.
It was toxic, but predictable. Workers could survive in that loop. And survival was enough.
Toyota brought a different loop. At NUMMI, workers agreed to UAW pay — but only if they worked Toyota’s way:
- Trigger: A signal in the process — a defect, a jam, a team need.
- Action: Follow the standard, pull the andon cord, solve as a team.
- Reward: Respect, pride, teamwork. The dopamine and oxytocin hits that come from belonging and winning together.
The Gate Called Predictive Adaptive Capacity
In my book Habit Engineering, I call this decision point Predictive Adaptive Capacity (PAC):
“Is it safe to try something new? Will the reward outweigh the risk?”
At GM, the answer was no. Pulling the cord wasn’t safe — it invited punishment. So PAC was low, and the learning loop never opened.
At NUMMI, the answer was yes. Pull the cord, and you got support. PAC was high, so the learning loop activated. And once learning starts, behavior changes.
Why Japan Mattered
Habits don’t stick in a week. They take six months to two years to really embed. That’s why Toyota did something radical: they sent Fremont workers to Japan for a year.
No lectures. No posters. Just the line. Their hands learned first. They pulled cords, solved problems, and were rewarded with help instead of blame. Slowly, the old muscle memory gave way to new habits.
When they returned to Fremont, those habits had weight. And when reinforced inside a Toyota-run environment, they stuck.
Why GM Couldn’t Copy It
Here’s the overlooked truth:
- Toyota controlled NUMMI’s management. Old Fremont supervisors weren’t brought back. Their cortisol-driven habits were too far gone.
- GM managers visited NUMMI — but they didn’t live it. They observed, then returned to Detroit unchanged.
- Toyota used NUMMI to train its own leaders. GM never built that pathway.
So while Toyota transferred habits back home, GM only took notes. And as one Toyota executive later said: “We showed them everything, and they still didn’t change.”
The reason was simple: GM didn’t need more knowledge. They needed new habits. And they had no safe place to build them.
What This Means for YouNUMMI wasn’t about copying tools. It wasn’t about waiting 30 years for culture to “emerge.” It was about engineering safe, rewarding habit loops long enough for them to stick.
If you want TPS-level transformation, the lessons are clear:
- You Don’t Need Japan, But You Do Need Loops
Don’t start with speeches about culture. Start with daily, repeatable loops that are safe and rewarding. - Expertise Comes from Safe Practice
Habits don’t change overnight. Build environments where it’s safe to try, fail, and learn for months — not days. - Create Transfer Pathways
Don’t just send managers to observe. Put them in habit-rich environments until they live the new loop. - Build PAC First
If people don’t feel safe to adapt, the old loop will always win. Safety is the gate to learning. - Hands → Habits → Belief → Culture
Culture doesn’t drive habits. Habits drive culture. NUMMI proved it.
Conclusion
NUMMI succeeded because Toyota didn’t start with mindset. They dismantled cortisol-driven loops and replaced them with dopamine-driven ones in a safe environment — long enough for new habits to stick.
That’s the hidden lesson most companies miss. And it’s the reason so many change efforts fail.
If you’ve ever wondered why your own Lean or TPS effort stalled — and what it takes to make change stick — that’s exactly what I explore in my book, Habit Engineering.
Because the real question isn’t whether your people “believe” in change. The question is: have you engineered the loops that make belief inevitable?
Proudly powered by Weebly