In 2006, I started this podcast. Jeff Liker was guest number 3, in August of that year. He has been back on seven more episodes since, which makes him one of the most frequent guests this show has ever had.
For the latest episode, I pulled clips from across those conversations going back almost twenty years. Listening through them again, what stood out was how much hasn't changed. The lean tools are better known now. There are more books and more case studies. The deeper thing Jeff was naming in 2006 – that companies want the words without the work – is the same thing he is still saying in 2026.
These aren't his greatest hits. They are the ideas that keep showing up.
Listen Now:
The Two Percent Problem
One of the first things Jeff said to me, and he has said versions of it in every conversation since, is that the gap between talking about lean and actually doing TPS is enormous.
Today I would say well over ninety percent are talking about it, and almost all of them have done something. They have done individual projects. They have hired consultants. They have done projects in individual plants. They have done Kaizen workshops. They have learned the words, and they have seen some results. The number that have deeply implemented the Toyota Production System as a system in their plants is probably below two percent.
What they have done is use individual tools in individual places. Occasionally there is a plant manager who really has a good feel for this, drives it through the whole plant, does a great job. They leave the company, and the plant reverts back to what it was. If you were to try to find a model supplier that really has TPS broadly across the plant, it is hard to find.
That gap shows up in a particular way. Companies decide they have mastered lean in their plants, and now they are ready to take it across the rest of the business. Jeff isn't buying it.
We got great success – now it is time to move to the enterprise level. Unfortunately, if you go into their plants, you see something different. They have not done it in their plants. They have done individual projects that have been successful, but their plants are far from what you see in the Toyota Production System. So they have superficially done it in their plants, and now they are ready to superficially do it throughout the rest of their enterprise.
How Long It Actually Takes
If you have heard a consultant pitch lean transformation in fourteen weeks – or twelve, or six, or fifty-two – this next clip is worth your time. Jeff describes what it takes when Toyota opens a brand new plant under what he calls ideal conditions.
When Toyota starts a new plant – like when they opened the one in Canada, or the one in San Antonio for the Tundra truck – they put about two years into developing and preparing the leaders to understand TPS. They send them to Japan. They work in Toyota plants in Japan. They are immersed in it. They hire people. They train them for months. They set them up in a completely lean environment with all the tools of lean.
If you ask Don Jackson, who was leading the San Antonio effort, how long it will take for the San Antonio plant to really understand lean and live it and follow the Toyota Way principles, he will say about three to five years. And those are under ideal conditions, where day one you have hired people from scratch and indoctrinated them into the Toyota Way. They have been to Japan. They have models. They are part of a network. They have a mother plant.
If under ideal conditions, with everything in place, it takes Toyota three to five years, I do not know how a company that has never done it before can do it in fourteen weeks.
I'm with him there.
Picking and Choosing
Companies sometimes say they want to build their own system, custom-fit to their culture. That sounds reasonable on its face. Jeff's response is sharper.
I have mixed feelings, because on the one hand, they should be doing that. They should be creating their own system. They should have ownership. On the other hand, what they often pick and choose are things that are designed to reinforce their current management system and their current philosophy, which is short-term focused, project focused, program focused. So all the best practices they put in place will be a bunch of individual programs that really will not change things very much.
If you take all the principles of the Toyota Way together, what you are trying to drive toward is to become a learning organization. Very few companies are learning organizations. Mostly they look at their company as a technical system where they shove in inputs and tools and people, and they want products out and they want to make money. They do not understand that to be successful long term, they have to create a learning community where people are willing to share problems and solve problems and get better every day.
The picking and choosing doesn't usually produce a stronger system. It produces a system shaped to leave the existing management philosophy alone.
What Cho Said Was Hardest
Jeff once interviewed Fujio Cho, who was president of Toyota at the time. Before that, Cho ran the Georgetown, Kentucky plant – the first Toyota assembly plant in the US. Jeff asked him what was hardest to teach the Americans about TPS. Cho's answer was not the tool I would have guessed.
One thing that stands out is the principle of stopping when you have a production problem – the andon principle. I had interviewed Fujio Cho, who at the time was the president of Toyota, and had been the president in Georgetown, Kentucky, where he taught the Americans about TPS in their first Toyota assembly plant in the US.
I asked him, what's the hardest thing to teach the Americans? He said the hardest thing was andon – to teach them that it is okay to stop when there is a problem and pull the andon cord. I asked him how he taught them. He said, I had to go to the floor every single day and say, please pull the cord if you have a problem.
That reflects a culture we have developed that says problems are to be hidden and not admitted. If you admit a problem, you are the problem. You must have caused it. If you are a good worker, you want to be able to solve the problem yourself. Whereas in the Toyota system, the whole system is based on people surfacing problems so that teams can solve them. And the first assumption is always that the problem is with the system, not the individual. It may be with the individual, but the first assumption is the system.
Jeff then connects andon to hansei, reflection.
Stopping the line, losing production, is counter to every instinct we have about producing every part we can every minute of the day. And that leads to the other principle – reflection and hansei and continuous improvement. In Toyota, they believe reflection is necessary for kaizen.
Whenever they do something, they have an opportunity to learn from what they have done and improve. In Japan, that is actually quite natural. In the US, it is quite unnatural. We are used to charging ahead and solving the next problem, fighting the next fire. The idea of stopping and saying, we just launched this plant, what have we learned, so the next time things will go better – we do not do that well at all.
What you see in Toyota is they do not believe you can implement the perfect lean system. They believe the essence of lean is that you implement a flawed system, you know it is flawed, and then over time, through continuous improvement, you make the system better and better. The lean tools are designed to surface problems so you can make the system better. We think we should be implementing the perfect lean system, and if there is a problem, it is because somebody did not do their job right.
The Non-Negotiables
Jeff co-authored a book called Toyota Culture with Mike Hoseus, who spent years inside Toyota. Together they laid out what they called the non-negotiables – the things you can't trade away without breaking the system. This one is about how Toyota thinks about its suppliers, and what happens when an American purchasing manager finds a cheaper part overseas.
A manager is a teacher – that is non-negotiable. And doing everything you can to create job security for people is non-negotiable. So a purchasing guy in America saying, hey, I can get a deal on this part by moving to India or China and save thirty percent – it will require shutting down one of our suppliers in America – that is not allowed. That is not permitted.
The response from the Japanese teacher (and the Japanese still play that role of being the sensei, the senior teachers) would be to ask: if you do this, what message will it send to all the rest of our suppliers who we have been claiming are partners in America? Isn't there a way to solve the problem that does not involve shutting down a supplier in America? They would see that solution as an easy quick fix, without really studying the problem and without seriously considering alternatives.
Don't Skip Hats
Jeff spent time at Toyota's plant in the United Kingdom and describes something they teach about roles and authority. They call it “don't skip hats.” The idea is simple. The discipline is harder.
In their training room they have a row of hats, and each hat has a different symbol on it. One says, I am a team member. One says, I am a team leader, and so on, depending on how many bars there are. They show an arrow from the team leader to the team member, from the group leader to the team leader, from the manager to the group leader. And they say, don't skip hats.
If a team member were to pull a cord, and a manager were to come down and ask the team member why they pulled it – that would be considered inappropriate. The manager should be asking the group leader. One of the things the manager is trying to determine is whether the group leader knows, and whether they have investigated enough to know. The group leader then uses that data to coach the team leader.
If the manager goes down to the floor and asks the worker, what happened and why – number one, they are usurping the authority of the team leader and group leader. Number two, they are not going to have the depth of understanding the team leader has, so they are likely to jump to conclusions. There may well be a reason, based on past history and various conditions, why the team leader and group leader reacted the way they did. The manager will not understand that, and they may mess things up.
A common issue when we are teaching managers in other companies is that they think the reason you go to the gemba is so you can solve the problems – as opposed to observe. You are a fly on the wall, and then you have to think and reflect on what you should do about this.
Five Whys vs Five Whos
One reason I wrote The Mistakes That Make Us is that the default instinct in most organizations is to find someone to blame. That happens, unfortunately, even in companies that would call themselves lean. Jeff has a way of putting this that I keep coming back to. He calls it the difference between the five whys and the five whos.
That comes from Dr. Deming, who said, drive out fear. He said ninety percent of problems are system problems, and they are the responsibility of management. He was not saying one hundred percent. There is some percent that really is somebody just was not paying attention. But he is saying in the vast majority of cases, it is the system, so let's start with the assumption it is the system. That became kind of religion within Toyota.
So they talk about the difference between the five whys and the five whos. The who is who is to blame. The five whys is what allowed this to happen. When you do the five whys, it is not unusual that any person's natural inclination is to blame the person. They might say, so and so dropped the nut, or so and so put on five nuts instead of six. And then you could say, what are you going to do about it? In Toyota, they would say, why did that happen?
They would not say, do not blame the person, that is not our policy, and start reading the rights act. They would just say, all right, fine, the person dropped the part – why did they drop it? Let's say they said, because they were not paying attention and did not follow the standard work. Again, you do not need to argue with that person. You need to say, why did they not follow the standard work?
Jeff also makes a point that I think gets lost in most five whys training. The goal isn't to keep asking why until you hit some ultimate metaphysical root cause. The goal is to find a controllable cause that will make an impact.
You do not want to find the root cause, but a controllable cause that will make an impact. If you keep asking why enough, you will get to things outside of your control – it was product development's fault, and that is too far. We are not going to see a change in product development until the next new model introduction. So what can we work on right now?
I have always joked (and I won't claim it was original) that if you ask why too many times, you end up blaming society, or Congress, or a meteor strike. Jeff agreed and added evolution to the list.
Twenty Years
Eight conversations across twenty years. Jeff has learned a lot. He has not changed his mind about the core things. I am not sure he needed to. The two percent problem he was describing in 2006 is still the two percent problem.
What I keep wondering is whether twenty years from now we will be having a different conversation, or another version of this same one.
