Shmoozing About Lean Leadership with Beau Keyte

Last week, Beau Keyte sent me a copy of Chapter 10 of his book The Complete Lean Enterprise, entitled Leading in the Future State, and we have been exchanging thoughts about it by email since. Many of my words below are already in my account of the Central Coast Lean Summit, where this conversation started; what is new here is the back-and-forth.

Michel Baudin: In management, PDCA sounds like the “Plan, Organize, Lead, and Control” (POLC) as taught in Business Schools as the functions of management. Of course, Plan, Organize and Control are more formal and easier to teach than Lead which, as a result, tends to get short shrift. Also, both PDCA and POLC can be interpreted as a sequence of separate tasks that the manager does alone, based on his or her infinite wisdom. But it can also be interpreted as dimensions of the work that needs to be done, and intertwined. Hoshin Planning, for example, is planning, but it involves the whole team and requires leadership. Conversely, leading a team requires thinking through the best way to do it in the circumstances and with the individuals involved, which is planning…

Beau Keyte
Beau Keyte

Beau Keyte: I see management as its own process needing continuous improvement, so I’m comfortable with PDCA as a learning system to understand how to manage and lead better.  My thought is that “Control” at the end of POLC seems to hold onto Command and Control.  I prefer to think of the final objective as “Develop.”

Michel Baudin:  Control may not be the most attractive part of the manager’s job, but it cannot be ignored. Its absence leads to events like:

  • In 1995, a junior trader in Singapore, Nick Leeson, was able to bankrupt Barings, the bank where J.P. Morgan had apprenticed.
  • In 2008, another junior trader in Paris, Jerôme Kerviel caused Société Générale $5B in losses.
  • The most charitable view of the 2015 VW emissions scandal is as a failure of control.
  • On a smaller scale, in machine shops, you have engineers in a hurry bypassing the engineering change management procedures and directly implementing changes to CNC programs, with occasionally disastrous results.
  • In addition to mistakes, you also encounter malicious or dishonest behaviors.
In one form or another, the managers have to exercise control, noticing discrepancies, and reacting appropriately, but it doesn’t mean managing by command-and-control. “Check” and “Control” both start with “C,” and have similar meanings.

Beau Keyte: I think it stilts the discussion to only one side demonstrating that the sky might fall when we trust people who report to us to do the right thing.  How should we trust leadership to do the right thing?

Michel Baudin: It’s all a matter of balance. My now grown son used to lie to me about brushing his teeth when he was five, and he now he blames me for taking his word. There is a fine line between being appropriately trusting and too trusting.

Beau Keyte: There were leaders inside the VW scandal (I think) and they didn’t do the right thing!

Michel Baudin: That’s why I was saying that attributing it to a failure of control was the most charitable view.

Beau Keyte:I have an issue with the assumption that control could have stopped any of these examples from occurring (where’s the data to support this assumption?)

Michel Baudin: I don’t know, but I would look at the procedures followed by investment banks that were not brankrupted by rogue traders, or the engineering change management practices of the best machine shops.

Beau Keyte: I am also concerned about the lack of discussion on the downside of this type of focus.  (e.g., the great stuff that was stifled by controls either in improved customer service, new heights of productivity, better market focus, etc.)  I would rather the argument not sound self-serving.

Michel Baudin: Yes, controls, and standards for that matter, can stifle creativity.

Beau Keyte: How would you propose a neutral position that would make people think?

Michel Baudin: There are many dimensions to this issue, including the selection and use of metrics. Controls meant to prevent the sky falling should be focused on limited areas, identified through experience, but relentless in these areas. For example, you don’t want to constrain the thinking of your product designers, but you are determined to prevent proprietary information leaking to competitors.

On another topic, I would want to address the issue of responsibility because, regardless of the extent of a team’s involvement in making decisions, responsibility for the outcome stays with the manager. For this reason, I advocate the sleep-at-night test to draw the line on delegation: “Can I sleep at night knowing I will be responsible for decision A made by team members X and Y?” If the answer is yes, you delegate. Otherwise, you say “We’ll make that decision together,” and it means you keep the final say. And the line between yes and no moves over time as the team matures.

Beau Keyte: I understand your point.  My thought is that each level of the organization needs to be accountable for solving problems.  My boss should be developing/coaching me on those problems (and performance) that I need to be responsible for and lose sleep over.  For those who report to me, it’s my job to allow them to learn, including learning through failure.  That doesn’t mean I would allow them to test a hypothesis which could do grave danger to the customer or company.  It just means that every level of the organization must find a way to let go of doing too much thinking for the people who report to them.  We have some really smart people reporting to us (unless we fail in our hiring and development process……)

Michel Baudin: With regards to managers injecting their own solution ideas in discussions, I think there is another issue besides ownership, in that their ideas are never assessed objectively. Most members will systematically agree with the boss, and a few make a point of asserting their independence by disagreeing systematically. Nobody assesses the content as if it came from a regular team member. It is particularly dangerous because managers tend to overestimate their technical skills. They may have been great engineers 15 years ago, but they are rusty and out of touch with technology.

Beau Keyte: I agree.  But I agree because the manager has no business trying to solve problems for those who report to him: there is no first hand knowledge of the context as he/she doesn’t do the work.  Too much thinking about something they know little about.  Their time is better spent solving their own problems.

Michel Baudin: I tend to view Lean Transformation more as the exploration of uncharted territory than the construction of a house. At the outset, a manager cannot have the information needed to draw a detailed vision of a future state. He or she can see until the next ridge, but not beyond.

Beau Keyte: I agree.  That’s what makes policy deployment (or some form of it) helpful: what do we need to accomplish this year, and where are the pain points in getting there?  And, what cross functional team needs to address this and at what level of the organization?  What’s everyone’s role in supporting this need?

Michel Baudin:  The opportunities come into focus as you go. I think the future of any business organization is uncharted territory regardless of whether you go Lean or not. What you need to do is endow the organization with the skills it needs to deal with the game changers that are certain to come its way, whether in the form of a 70% drop in oil price or the introduction by Apple of an “i” version of its product.

Beau Keyte: Great point.  Teaching them how to think and act differently at every level of the organization will give them the adaptability to survive lots of market change.