I have long been concerned about organizations that struggle with Lean transformation and the impact on the company, its employees, and other stakeholders. People’s lives and livelihoods depend upon Lean, so it’s no small matter when Lean succeeds or fails. This has been a major focus of my work for over 20 years because I wanted to know the cause-and-effect relationships for both successful and failed Lean transformations. That is why I divided my research and writing between these two very interesting poles.
John Shook has put forth the idea that if we undergo a change in mindset from “implementing” Lean to doing “experiments,” then “the meaning of failure and success changes.” His reasoning suggests that any failure is a success if there is good reflection and learning that comes from it, to achieve a better result next time. But for organizations whose Lean transformation process failed, there usually is no “next time.”
For something as consequential as Lean transformation – for the business and for its employees – we should want to know far more about the causes of success and failure than that which may come from reflection alone. We should want rigorous analysis of both success and failure, which are complementary to one another, to provide a more complete picture. It is in everyone’s interest to help organizations do better by using a structured problem-solving process such as my A4 method and creating process-at-a-glance charts for Lean transformation (as my students have done in my university course on Lean leadership).
While the meaning of failure and success may change for Lean, thanks to John’s reflection, it does not change for managers. Whether we like it or not, business leaders have long had a very clear and specific understanding of success and failure, and reflection and learning are not part of it. Perhaps, in time, they can be taught otherwise.
As we all know, Toyota’s production system (TPS) is notable in large part because it applies the scientific method – and derivatives such as kaizen, PDCA, A3 reports, etc. – to management, in order to achieve flow. Therefore, the importance of experimentation to as a means to create TPS and beyond has long been abundantly clear. (Note: In the English translations of his books, Taiichi Ohno often says “implementing” TPS, and frequently used the terms “experiments,” and “trial-and-error”).
An experiment relies on a method. A method (such as a research method in science) can be poor and result in failure. Sound methods yield sound results, whatever the result may be. And sound methods are reusable by others. In the context of Lean transformation, the method is poor and the result is failure. So, problems with the method – Lean “implementation” – need to be better understood, including the specific modes of failure, before moving on to “experiments.” Furthermore, changing the mindset to “experiments” may do little or nothing to reduce Lean transformation failures. And, more time will slip by without understanding the true nature of the problems or identifying effective countermeasures.
The way I see it, several long-standing assumptions must be questioned:
- Organizations must start from zero.
Toyota had to start from nearly zero to create their flow production system. They surely benefited from the work of others before them, such as Ford Motor Company and Morris Motors Ltd. But after 44 years of practice and study of TPS and Lean, why must every organization start from zero? Why must every organization repeat the same mistakes? Why do we let this happen?
Collectively, we have learned enough over the last 44 years to specify Lean transformation processes that that have a high probability of success and specify those Lean transformation processes that are sure to fail. The good processes are reusable. We should encourage the former and strongly discourage use of the latter.
- Every organization is different.
Is that so? What most of us actually find is that organizations are far more similar than different. They all process material and information using the batch-and-queue method and would benefit from flow. People are different, but t-shirts don’t come in 7.5 billion sizes. They come small, medium, large, and extra-large for men and women. By saying every organization is unique and that the only starting point is “experiments” likely turns off a most managers. What managers want is help from internal or external kaizen consultants to convert processes from batch to limited (supermarkets) or continuous flow. What managers then have to accept is the constant experimentation that must be performed by employees to improve flow, innovate, and adapt to changing conditions.
After 44 years of studying Toyota and others, and lots of books and academic papers, we know what works and what doesn’t. What good is all that work if it is not used? Respect the work; use it! The amount of experimentation to achieve certain key outcomes is less than is commonly thought, while, of course, the need for experimentation never goes away because change is ever-present.
- A more rigorous, deliberate scientific approach to Lean transformation appeals to people.
The public gained awareness of a more deliberate, scientific approach to Lean with the publication of Mike Rother’s book, Toyota Kata, in 2009. Continuing the theme of changing mindsets to “experiments” could be a good thing, but maybe not. I worry that moving Lean in the direction of hypotheses and experiments – more overtly science-based – could give some people the impression that Lean is exclusive, intellectual, esoteric, or elitist. This will this narrow the audience for Lean to those who comprehend science or otherwise limit the appeal of Lean.
Today’s populist movement is generally anti-science. Is it good thinking or good timing to position Lean as science? Are business leaders eager to promote Lean as scientific experiments? Is intellectual rigor something that drives executives? No, No, and No.
In the early days, Shingijutsu did not explicitly teach us Toyota kata, experiments, PDCA, A3 reports, value stream maps, etc. They taught us kaizen. That was how we learned to see waste, unevenness, and unreasonableness, stagnation, and other abnormal conditions, and learned how to eliminate it via “trystorming.” Yes, trystorming is experimenting, but sensei did not convey a message of science.
Kaizen was perceived differently and carried meaning that was acceptable to a wide audience – engagement, teamwork, idea generation (flood of idea vs. few ideas from brainstorming), making things, and realizing impossible dreams. People responded positively to sensei’s effort to uncover their hidden desire to improve and ability to do great things quickly. Quickly! And we were pretty successful – for a time, as is usually the case.
The formalisms that have become part of Lean over the last 20 years did not exist. In my view, these formalisms have had the effect of creating a bigger gap between managers’ and employees’ current job routines and their future job routines. Bigger gaps are harder to close. And people lose interest when gaps are too big.
- Failure does not exist; success is reflection and learning (and better luck next time).
When managers lay off employees after making Lean improvements, is that nothing more than an experiment that did not work? Is leadership no longer responsible for such zero-sum outcomes, which ruin people’s work and family lives, and also stall or kill Lean transformation? Experiments absent of guiding principles, such as “Respect for People,” are bound to fail and cause harm to people.
Executives don’t want to hear that their Lean transformation was just an experiment that didn’t work, and success is actually determined by how well they reflected on it and what they learned, for next time. Employees don’t want to hear that either. If Lean is defensively characterized as: “Success = Success and Failure = Success,” does that expand or lower its credibility as a method for improvement? Can it withstand the onslaught of criticism from the Lean haters?
- Executives must take what we give them because we know better.
The hypothesis-experiment formalism will likely elicit criticism from executives:
- “Experiments take time. We don’t have time.”
- “I care about implementation – fast implementation! I don’t care about experiments. That’s too slow.”
- “I want implementation results, not a bunch of experiments.”
And criticism from the managers that report to the executives:
- “I don’t have time for experiments! I have to deliver results.”
- “My boss wants results, not experiments.”
- “My boss told me to stop doing experiments and get to work.”
To them, business is economics, not science. And economics is powerfully anti-Lean.
What will your response be, John Shook? Executives prefer implementation to experimentation. A more overt association with science could make Lean less appealing to top decision-makers. It seems likely they would instead opt for something different to get the results they need. Plus, the mindset for experimentation is rare – even among the CEOs who used to be scientists.
Over the last 30 years, executives have cared more about the present than the future. That’s why they opted for Fake Lean instead of Real Lean, and the esteemed leaders of the Lean movement let them do it without public rebuke or offering what executives might has seen as viable alternatives.
Let’s think about this: What do executives want? Assuming they want Lean, they want faster, not slower, Lean transformation. The demand is for faster Lean transformation. Why haven’t we given that to them? What else do executives want? They want a Lean transformation process that requires less, perhaps minimal or even zero, senior leadership engagement. There is demand for that. Why haven’t we given that to them?
Perhaps it is because we do not know enough about Lean success and failure due to a lack of structured problem-solving. Or, maybe we know plenty about both but have not made the effort to present practical processes that work and those that do not. Why haven’t we engaged in trystorming to create several new methods to meet these and other demands?
These are two real-world demands that challenge our collective thinking. We must accept the challenge if Lean is to survive.
• • • • •
Some organizations fail in their Lean transformation process at the outset, others fail later sometime in the process. Together, we have the knowledge to figure out what is going on in either case and create better processes that other organizations can use to achieve non-zero sum outcomes and long-term survival.
But for some reason, people are left to struggle – as if that is the way things should be. This cannot not be the case after 44 years of TPS and Lean practice and study. If it is, it proves that Lean is not about learning.
John Shook’s closing comment was: