Dear Gemba Coach,
In many classic lean stories I hear, the sensei comes across as completely obnoxious. Is that really necessary?
“Are you responsible for this plant? Kawabe acknowledged that he was. Then Ohno roared, “This operation is a disgrace. You are completely incompetent. Yamamoto-san [the President], fire this man immediately!”
That was the first story I ever read about Taiichi Ohno. I had been studying TPS on the shop floor for my PhD and devoured Lean Thinking as soon as it came out. The Toyota people I was observing at a supplier had never mentioned Ohno, and certainly didn’t behave like this but the Japanese sensei did have that reputation and, indeed, when this could be the style of CEO/sensei visits. In Lean Thinking, the story goes on to say how Kawabe went to lead the total company transformation – but, as a sociology doctoral student studying mental models, I really wondered what I had gotten myself into.
Fast forward to present times, on the gemba. This is a digital startup in a financial services company. The CEO and the sensei have just had a great discussion with the IT VP about how to better control the flow of changes to the website to avoid mishaps in production. Starting from a glitch, they’d worked through the workflow and discussed possible causes and places to start to improve autoquality. All interesting, everyone engaged and in a good mood.
The next obvious step back on the gemba is the design team that generates many of these change requests (one of them – in such a company, change requests arrive from several sources, which is part of the difficulty). The lean program manager has argued hard for a visualization of the changes emerging from the service design team. This team thinks of things to make the service easier or more attractive to the customer, which then have to be coded in the website.
The sensei picks up the card, reads it, and starts being obnoxious: “I can’t read this, it’s too small, it’s too verbose, it doesn’t tell me anything clear – it doesn’t help.”
The point the sensei is trying to get across is that the card’s purpose is to make the change clear (what and why), as well as how the change should be implemented (where exactly), so that all other teams involved understand what this is about and can both contribute opinions about difficulty and help in implementing these changes – both reasonable support and pushback. The wider purpose is to strengthen teamwork across the workflow – and avoid the kind of glitch they just witnessed.
The team leader pushes back on every point. Not only is the card perfectly clear for anyone who needs to understand (look, this is specialist work), but why do they ask us to handle these cards in the first place (we have no time for this extra work)? The CEO says mildly that every time they’ve visualized a process better, they’ve had visible gains. The guy is unconvinced and sticks to his guns. The sensei starts again, explaining to the whole team why these cards are important: They are not logging changes on the wall as opposed to on the computer. They are about explaining clearly the changes to the rest of the company.
One person in the team agrees, saying, “Yes, I can see the card is unclear, it makes sense we could work on that.” But then the team leader jumps in again without letting her finish and returns to his original gripe about no time for this stuff.
Then it all goes south. The CEO withdraws looking angry. The sensei becomes ironic and tries to go around the leader and speak with the team, and then gives up and walks away saying “as you wish,” dismissively. Everyone is being a jerk.
The aftermath is no simpler. The team is all upset and incapable of focusing back on the work. CEO and sensei had gone to the team to follow up on a problem investigation – which now has become a people problem: did the team leader get surprised and react badly – or is he always like this? The sensei is difficult, but then we knew that.
And here’s the rub of it. The basic learning theory underlying lean on the gemba, on-the-job learning is action learning:
On the job Learning = Programmed Knowledge + Insight Questioning
In lean terms:
Learning = (working with standards so all goes well) + (investigate one waste elimination topic)
Team leaders are expected to do both:
- 99%: Solve problems every day so that standards are maintained in varying conditions and the current way of satisfying customers is upheld, the flow of work is steady, and the team knows what it’s doing.
- 1%: Tackle one work process improvement with the whole team so that they reflect on their work and come up with insights and initiatives.
On the one hand, we need to be boring and follow routines, solving problems to go back to the routine. On the other hand, we need to break routines one by one by following the most interesting path.
Routines are necessary to find a comfort zone. One challenging/interesting question is necessary to be stretched some. Too much stretch and people fall into the panic zone, where they behave at their worse.
The senseis I’ve seen are rarely unpleasant on the gemba. They don’t mock, insult, threaten, make bawdy jokes, or act rudely. But they also can be, each in their own way. When the CEO/sensei relationship is established in the company, and the sensei points to a problem, with either 1) visualizing better a process and 2) looking more deeply into the problem revealed or 3) working more closely together, most people accept it and move to do following where the sensei is pointing at rather than staring at the finger. But then when someone will not accept to look into the topic pointed at by the sensei, the sensei’s choice is to move away or push harder. In practice it’s usually 1) push harder, 2) move away and… 3) discuss with the CEO why that person didn’t want to look deeper into the problem. This is an endless discussion between sensei and CEO that revolves around skill (the person is afraid of not being able to handle the problem uncovered), will (the person has no intention to improve) or weirder (they’re hiding something).
Senseis, obviously, are not superhuman and quite capable of being obnoxious in their own right, and of getting an issue wrong. And they do. But that’s not the point. A sensei is like a compass – their years of experience and track record make their instincts worthwhile. When the compass points North, the compass need not know why it does so for you to follow its direction. Don’t have a dog and bark in its place. Don’t bring a sensei on the gemba and not explore where she is pointing. The dog might be barking for no visible reason – or they might have picked up something you haven’t seen. The sensei might be making a fuss because they got up on the wrong foot this morning, or they might be picking up an important thread.
As you say, classic lean stories are about the sensei being outrageous (i.e. nasty to the people going through it in situ) and thus uncovering a spectacular path for progress no one had hitherto seen. These stories make up the lean myth, and how much of lean tradition is transmitted. Of course, no one passes on the stories when the sensei was nasty and there was no learning opportunity there in the end, or the many, many more cases when nothing happened because CEO, sensei, local manager and team just worked without fuss.
The key factor is psychological safety – and is much more a CEO issue than a sensei issue. In the previous case, it is quite clear that the CEO has never fired or demoted anyone over a bad gemba walk. It is known that the CEO spends more time with people where gemba walks are fun. However, it is also known that some people have left because they didn’t like these gemba walks. Tricky. Psychological safety is the key ingredient that makes it safe to be caught off guard and behave terribly on the spot.
When problems are really tough, the mental effort needed to face them tends to get people emotionally screwed up as well, and so they rarely appear in their best light – if they’re fully engaged, they’re skirting the stretch/panic line and can that be awkward. The CEO is the person who then maintains a spirit of psychological safety in cooling conflict when it rises up, consoling the loser, and generally showing that as long as people are authentic and honest, no one will hold it against anyone that tempers rose and things were said.
Still, psychological safety is always fragile, because there remains a mysterious line which, once crossed, can’t be walked back from. Once the safety is gone, the situation becomes toxic and everyone loses.
No easy answers. Challenging people to: 1) better visualization, 2) deeper investigation, 3) clearer more thoughtful experiments, 4) more teamwork is the lean method. A successful gemba walk is one in which the CEO has learned something and has been able to compliment people on their efforts. All goes well as long as everyone you meet is cool about it. Then at some point, for whatever reason, the gemba walk goes wrong. This is clearly to be avoided, it’s also unavoidable, so I have no generic answer to these situations other than: it depends – treat them case-by-case, and don’t hold it against them when people (or you) find themselves less than perfect. Move on to the next gemba and try to do better next time.