One reason why GM fell to Toyota.

A couple of weeks ago, the Planet Money podcast aired an episode on Studs Terkel. Recently archivists found the interview tapes that formed the basis of his 1974 book, Working.

I’ve been planning to write about one of the interviews for a couple of weeks now, but I couldn’t quite bring it into focus. Today’s Gemba Academy blog post by Jon Miller, about Toyota Exec VP Mitsuru Kawai, helped clarify my thoughts by highlighting some fundamental differences between the way GM managed its employees in the 1970s, and the way Toyota deals with them today.  

Terkel interviewed Gary Bryner (starting at 8:13 here), a line worker at the GM Vega plant in Lordstown. Around the time of this interview, GM had installed Unimate robots on the line, increasing production from 60 cars per hour to 101 cars per hour, and making it the fastest line in the world. (Note: 101 cars per hour? I’m not an industrial engineer, but that seems insanely fast. Granted that cars were simpler then and perhaps easier to assemble, but it’s hard to imagine that workers could do a good job with only 36 seconds per car.)

Terkel: Yeah. So what happened to the guys in the plant that are working there now?

Bryner: It’s a funny thing. You know, when they revamped the plant [by installing the Unimate robots], they’d try to take every movement out of the guy’s day so that he could conserve seconds and time, so they could make him more efficient, more productive. 

GM’s reason for trying to be more efficient is if they could take one second and save a second on each guy’s effort, they would over a year make a million dollars. You know, they used the stopwatches. And they say, “Look, we know from experience that it takes so many seconds to walk from here to there. We know that it takes so many seconds to shoot that screw. We know the gun turns so fast with the screw so long and the hole so deep. We know how long it takes. And that’s what that guy’s going to do.” 

By contrast, here’s how Jon Miller describes Toyota Executive VP Mitsuru Kawai’s comments about Toyota’s New Global Architecture (TNGA):

Interviewer: In other words, TNGA is about the standardization of parts?

Kawai: No. In the first place, TNGA is a much larger effort. Simply put, it is “involvement by everyone.” 

He described a product development line in which the engine assemblers are over sixty years old. Their job is to do the work and complain, “It’s heavy! It’s difficult! It’s not error-proof!” so that the young team leaders and engineers use their brains to make the work lighter, easier and error-proof.

You can see the profound difference in approach here: at GM in the 1970s, managers and engineering experts define how, and how fast, work should be done. In Kawai’s description of TNGA, the line workers play an integral role in creating a better production process and design. 

On a deeper level, GM’s approach of entirely separating thinking (product design and production engineering) from doing (physical assembly), was disrespectful and inhumane. Here’s Bryner again, complaining to Terkel about the way their work was dictated by engineers with time-motion studies:

Bryner: Our argument has always been, “You know, that’s mechanical. That’s not human. Look, we tire. We sweat. We have hangovers. We have upset stomachs. We have feelings, emotions. And we’re not about to be placed in the category of a machine. 

Terkel: This is something new, isn’t it? The workers in the plant – they feel that they have a right to determine the nature of their work, too?

Bryner: We do now. We have some kind of pride being able to stand up to the giant, General Motors Corporation, and say, “Look, this is what I think is fair, and I’m willing to fight to show you that it’s fair. I just think they want to be able to be treated with dignity and some respect. And you know, that’s not asking a hell of a lot.”

The situation—and the expectation—at a Toyota plant today are quite different. In another interview that Jon Miller relates, Kawai expresses his joy that workers are actively engaged in designing and improving the assembly line: 

I am very happy that things are changing every day. On every walk, I notice “That diagram has changed,” or “They were able to shift that work,” or “Now the robot is doing that.” The fact that there are changes each time I visit tells me that people are doing kaizen. Everyone is trying to make good products at lower cost. This shows me that the gemba is alive. It is changing every day. It is reborn every day. I think this is the power of the gemba at Toyota.

I’ve never worked on an automobile assembly line, but I’m pretty sure that even at Toyota, it’s not like being on holiday. However, the stark difference between GM’s and Toyota’s management philosophy in the 1970s goes a long way towards explaining why the Detroit colossus was toppled by the Japanese upstart. I only hope that GM’s managerial mindset is different today. 



Your Lean Six Sigma Belt Program *IS* The Problem

I visited a company a few weeks ago that asks all of their employees to do a green belt project. It’s not mandatory, but completion of a project is part of their annual review. Not surprisingly, the management boasts that nearly everyone does a project. 

You know how many people do a second project? Less than 5%. 

This company is doing okay, but they definitely don’t have a culture of continuous improvement. Plenty of green belts, sure. But there’s no evidence of a continuous improvement mindset among the workers. Which is what the senior leadership wants, and why they created a Lean Six Sigma belt program in the first place. 

Read the rest of this post at Industry Week.



How "scripting the plays" can help launch your lean initiative.

My friend Bob is struggling to get a lean transformation started at his new company. He’s knowledgeable and experienced, but he’s been unable to get traction at the main plant. He’s tried everything—teaching intro classes to lean principles, taking people on waste walks, putting them in Ohno Circles, leading 5S events, practicing Toyota Kata—nothing has worked. He’s made as much progress as a piece of congressional legislation over the past few years.  

Bob believes that one of the primary causes for the lack of progress is the shortage of managerial and supervisory experience. Most of his plant leadership team was recently promoted from front-line roles. As a result, it’s difficult for them to break the habit of solving problems themselves. When something goes wrong, they immediately dive in and fix it, rather than coaching workers to understand root cause and deal with it at their level. And that makes it nearly impossible for Bob to get the managers and supervisors to focus on continuous improvement because they’re constantly distracted by the latest problem.


But of course, the more fires they douse, the more fires break out. Unless they develop front-line operators’ problem solving capability, they’re sowing the seeds of their own demise

So what to do?

Bill Walsh, the legendary football coach who led the San Francisco 49ers to three Super Bowl championships in eight years, was famous for scripting the first 20-25 plays of each game for his team. (This is now a habit for many NFL coaches.) Pre-determining the early phases of the game provided two major benefits. First, the ability to rehearse the exact sequence of plays during practice enabled the team to execute more cleanly during the game. Second, he could test how the opponent responded to certain plays and formations, helping him implement more effective countermeasures later in the game. 

It occurs to me that Bob might be able to “script the plays” for the managers and supervisors, which might bring stability to their work, and teach them how to coach, rather than solve, shop floor problems.

“Scripting the play” goes beyond leader standard work. It’s not just blocking out time for a gemba walk or a meeting. Rather, it would be a precise agenda for the first 30 or 45 minutes of each day: walk the floor along a pre-set route; ask a set of pre-determined questions; take 10 minutes for prescribed reflection with Bob. Only if a literal (not metaphorical) fire breaks out would they be allowed to vary from the script. 

Sure, the company could invest in leadership training. But classroom lectures, readings, and exercises are, at best, difficult to transplant to actual shop floor interactions, and at worst, a waste of time. Scripting the play might be the wedge, or the on-ramp, needed to instill new habits and help the managers and supervisors move from an operator mindset to a manager mindset. 

Have you tried anything like this? Let me know. And I’ll keep you posted on how this turns out. 



The road to hell is paved with metrics.

Michael Harris and Bill Tayler wrote a terrific article in the new Harvard Business Review on “metrics surrogation”—the tendency for people to mentally replace their business strategy with metrics. Although these metrics are supposed to reflect the company’s progress towards its goals, the focus on metrics can destroy an organization.

Wells Fargo’s “”Eight is great” initiative, which had the company measuring daily cross-sell numbers, caused employees to focus on increasing that number—not necessarily deepening the bank’s relationship with its customers. 

The Atlanta public schools cheating scandal, in which teachers received performance bonuses based on students’ scores on standardized tests, provided incentive for teachers and principals to erase wrong answers and replace them with correct ones—not necessarily improving students’ mastery of the underlying material. 

And, of course, call centers in which employees are evaluated on how fast they’re able to get off the phone motivates them to cut calls short in a variety of ways—not necessarily by solving the customers’ problems. 

In all these cases, the underlying performance or goal the organization is striving for has been replaced by the measurement of that performance. All too commonly, that leads to behavior that’s dysfunctional at best and illegal at worst. 

The authors recommend three countermeasures for this problem:  

  1. Get the people responsible for implementing strategy to help formulate it.

  2. Loosen the link between metrics and incentives.

  3. Use multiple metrics.

All of which is valid, I guess, but I think it adds unnecessary complexity to dealing with the problem. 

W. Edwards Deming had a much simpler way to avoid surrogation (even if he didn’t call it that). In his classic, understated way, he’d simply ask:

By what method?

In other words, what’s the method for improving performance? With that simple question, people would be forced to articulate the process they’ll use to reach the organizational goal. In the case of Wells Fargo, “illegally opening accounts for customers” clearly isn’t kosher. But “inviting customers to a local branch for a complimentary financial health checkup” might work. The Atlanta teachers wouldn’t announce that they were planning on fudging test scores. But “holding after school remedial instruction” might have done the trick. And call centers could provide workers with more technical training, or give them more latitude and authority to solve customer problems based on their own judgment. 

Keep it simple: By what method?



Start small. Move fast.

An improvement team at my client was struggling to get started on their experiment. They spent nearly a month interviewing people in multiple functions, trying to understand all the possible permutations of the current condition. They discovered that it’s really, really complicated. 

And that made it difficult to start improving the process. Trying to get all the pieces in place for their first experiment was laborious—so many moving pieces, so many people to include, so many variations to consider. 

No surprise there. Office processes are notoriously intertwined. You’re about as likely to find a problem confined to a single functional silo as you are a copy of the National Enquirer in the Library of Congress.  

Which is why it’s good to start improvement projects with small steps. Mike Rother suggests that initial Toyota Kata experiments should be sized for completion in about two to three weeks. Any more than that, and the complexity of the change outweighs the improvement skill level of the people making the change. 

Equally important, if the experiment goes on for too long, it runs the risk of becoming yet another time-sucking, dispiriting corporate project with undefined benefits in the indefinite future. That’s no way to get people excited about improvement.

Improvement Size 2x2.png

People are animals. They crave rewards as much as your dog wants a Milk Bone. As Teresa Amabile and Steven Kramer show in their book The Progress Principle, the gratification that comes from forward progress—no matter how small—is a powerfully motivating reward in and of itself.  When an experiment is too big and takes too long, people don’t get that emotional reward. And that makes the long continuous improvement journey feel like a life sentence in a salt mine rather than an inspiring walk in broad and sunlit uplands

This improvement team has just recalibrated. The process they’re trying to improve is still complex, crossing over multiple functional silos. But now they’re just taking a small piece of it, and are expecting results in a couple of weeks. 

They’re very excited. 

Start small. Move fast. 



An Unanticipated Benefit from Toyota Kata

I’m using Toyota Kata to help a client shorten its lead time for product design/delivery. This company sells primarily to department stores and mass merchants, a distribution channel that’s facing incredible pressure to shorten lead times—no one wants to guess six months out what’s going to sell, and inevitably be stuck with a bunch of closeout inventory. 

I have three teams working on experiments. As you’d expect with the kata approach, some of the experiments have yielded good results. Many more of them went nowhere. 

But the happy surprise for me (and the teams) was the improvement in cross-functional communication. The company is very siloed. Historically, people in one silo haven’t talked much to people in other silos, unless it’s to complain about something that’s gone wrong or to find out when something was going to get done. 

The experiments they’ve undertaken has forced the improvement teams to talk with people upstream and downstream from their silos. And that communication has really changed morale. It’s not all sweetness and light, of course, but there’s a greater appreciation on all sides for how their work affects others—and how difficult their jobs can be. 

We still have a long, long way to go, but the increased, and more productive conversations have already made a difference in attitude. 


1 Comment

How Scientific Thinking Won the Women’s World Cup Title

In the glow of the US women’s national soccer team’s World Cup title—and the long record of success the team has had over the years—it’s easy to forget that victory didn’t always seem quite so inevitable. Although the team dominated and won the World Cup in 2015, a year later it lost to Sweden in the quarterfinals of the Olympics.

Coach Jill Ellis noticed that Sweden employed a fundamentally different strategy than she expected. To add to the problem, many of her key players were getting older. She needed to make changes. But what changes?

Scientific thinking (Plan-Do-Study-Adjust, or PDSA) teaches us to experiment our way to improvements, because it’s impossible to know in advance what will work. Ellis stated publicly that she was going to put the team through an evolution that she felt necessary to win the 2019 World Cup. As Sports Illustrated reports, that’s precisely what Ellis did:

vowing to unlock more creativity in the attack, Ellis launched a period of experimentation (with formations and new players) that proved an old adage: Real change can be an ugly and uncomfortable process long before it becomes glorious.

Not all experiments are successful the first time. At the SheBelieves Cup the following year, the team notched two draws and only one win. The nadir was its game against France: the US team went down 2-0 after nine minutes to France and ultimately lost 3-0.

Critics assailed Ellis for not knowing what she was doing, and making the most talented team in the world look bad. After all, look at the lousy results.

But that’s precisely the point of experimentation—it’s to see what doesn’t work, so you can figure out what might work. That’s what happens on Mythbusters, except no one attacks those guys for “failing” when their initial attempt doesn’t work out. It’s the constant experimentation that makes the TV show interesting and enables them to get where they want to go (usually, blowing something up). Mike Rother captured this idea in a Toyota Kata slide:


By the following year, Ellis settled on an unconventional 4-3-3 lineup (four defenders, three midfielders and three forwards). She used different personnel from the Olympic team, including Rose LaVelle, who became one of the breakout stars in France.

In 2018, the team was undefeated.

It 2019 it won the World Cup.


1 Comment

1 Comment

"Process Mining"? Sounds Like a Waste of Time.

You know what the problem is with your lean initiative? You’re not doing any process mining. 

Yup, that’s right. You’re missing the boat on “process mining,” the latest improvement breakthrough that will catapult your firm to the top of your industry. According to a new HBR article, process mining will “revitalize process management in firms where it has lain fallow for years.” 

The authors state that companies adopting business process reengineering are so focused on the future state that they ignore a thorough analysis of the current state. Conversely, companies that adopt an incremental improvement approach

tend to spend too much time on analyzing the “as-is.” In addition, their current process analysis is frequently based on interviews and sticky notes, which executives sometimes regard as overly subjective and treat with justifiable skepticism.

I hardly know what to say about this argument. How do you define “too much time”? If you don’t know where you are and what’s actually happening, how do you know what to change? What kind of current state analysis only has interviews with no observational data? And are executives really skeptical of sticky notes? I could be wrong—I’ve never shown a hand-drawn process map to the CEO of a Fortune 10 company. Maybe executives at that pay scale really do have an issue with sticky notes. But you could always transfer the notes to Visio or Smartsheet or iGrafx if your CEO is allergic to Post-Its.  

Fortunately, now there’s a host of companies providing process mining software that will eliminate all these problems and set you on a smooth glide path to success. The software captures information and transactions as they flow through the company, and makes visible how long that (computer-mediated) work takes. 

For example, the global chemical company Chermours used process mining to analyze their order-to-cash process:

It took the process mining effort four months to uncover how the actual process was performing (not just what the ERP documentation stated). It made the entire process visible and revealed some glaring issues. Credit holds was one such issue, as process mining exposed that strategic customers were sometimes placed on credit hold needlessly to enable manual steps in the O2C process.

Look, I know that Chermours is a large company—4000 customers in 130 countries. But still. It took four months and an expensive piece of software to find out that strategic customers were needlessly placed on credit hold occasionally??? They couldn’t have figured that out in two days by going to the gemba, watching the process, and actually, you know, talking to the employees in the credit department? I’ll bet you dinner that when the CIO and CFO announced this Copernican insight to the company, there were a bunch of $20/hour front-line workers doing a facepalm.

Another great sales point for the software—the opportunity to anoint a class of high priests with specialized knowledge who take responsibility for improvement. At ABB, 

A small group coordinates the process mining effort in the head office, and as much as 80% of the process mining work is done by Quality & Operations personnel at the business unit level as part of ABB’s continuous improvement program.

Gee, that sounds terrific. Place responsibility for improvement in the hands of a few people who have the special skills and training to operate the software, and let them fix the problems. Leaving aside the fact that you’re ignoring the accumulated knowledge and wisdom of the front line employees, you’re setting yourself up for an improvement bottleneck and a nice long queue when the high priests of process mining can’t work with you for another six months. Continuous improvement? Discontinuous improvement is more like it. 

And finally, one last benefit of process mining: 

Process mining also contributes to reducing non-value-add activities and eliminating manual reporting efforts.

I have no doubt that it does make work more efficient. And so does close observation of the work with standard work combination tables. But that would only cost about three cents per page, which isn’t nearly as exciting as investing in a giant piece of expensive technology.

Look—I’m sure that sophisticated software can play an important role in process improvement, especially when you’re dealing with highly complex processes. But relying on software at the expense of direct observation is insane. 

I spent two days watching the warranty process at one of my clients. I saw how one person was slower than another because at 5’0” tall, she couldn’t reach the shipping boxes where they were stored, and she couldn’t access the database at the same time as the warranty lead. I’m sure that process mining software would have revealed precisely how much slower she was than her colleagues, but that would have completely missed the point.

Go see. Ask why. Show respect. Before you start mining. 

1 Comment


Adding Technology Isn't Always the Answer

It’s been a long time since I blogged about the inanity of multitasking in a cognitively demanding environment, but my friend Mark Graban sent me an article that reminded me of how important this topic is. 

According to a new study by Penn Medicine and Johns Hopkins, first year physicians spend 87% of their time on indirect patient care, half of which is consumed by the various electronic medical record systems. Worse, they spend only about three hours per workday on direct patient care, and they multitask during a large chunk of that time. 

I’m not a physician (and I don’t play one on TV), but I’m pretty sure that spending 87% of your time away from the customerpatient can’t be any good. To be sure, there’s certainly value—doctors need to do research, examine lab results, evaluate patient and family history, etc.—but 87%? And I know there’s incidental work that must be done in order to ensure high quality care and compliance with the thicket of regulations that hospitals and physicians labor under. 

But still: 87%? It’s hard to imagine a scenario in which 87% is the appropriate amount of time to spend away from the patients to whom caregivers devote their lives. And I can’t think of a single doctor (including my wife) who has said, “Yeah, I spend way too much time with sick people. There’s nothing better than a few hours in the privacy of my office messing around with the EMR.”

The multitasking issue is perhaps even more alarming. The time that the physicians have in direct contact with patients is so limited, you’d think that they’d want to maximize its effectiveness. But by multitasking, they compromise the benefits of the face-to-face interaction. The research on multitasking is voluminous and unequivocal): it simply doesn't work. Whether you’re a student taking notes, a businessperson reading email while in a meeting, or a doctor listening to a patient, if you’re splitting your attention between two cognitively demanding activities, you’re not going to do either of them well. When you think about the importance of using all your senses to really engage with a patient to deeply understand the problem, well, multitasking is a recipe for misdiagnosis, medical errors, and at the most basic human level, just plain hurt feelings. (“Why isn’t the doctor looking at me and paying close attention?”) 

The fact that the physicians are coordinating care or updating medical records for about 25% of the time they spend with patients reminds me of a story that my friend, Roger Chen, former head of CI at Martin Memorial Health System told me. Years ago, MMHS invested heavily in computers on wheels (“COWS”) for the patient rooms to make it easier for the doctors to enter information while they were with their patients. Unfortunately for MMHS’s finances, it turned out that patients hated the COWS—they felt that the doctors weren’t really paying attention to them when they were multitasking (listening and typing). So they scrapped all of them. 

We spend a lot of time in the lean community talking about value added activity, incidental work, and waste. Many hospitals are beginning to turn that lens on physician activity. It seems like a good idea to bring that mindset to the training of new doctors as well.



Better Managers Isn't Enough for Higher Engagement

(Note: this article first appeared in Industry Week.)

recent article by Sam Walker in the Wall Street Journal argues that better managers are the key to delivering better results.

Walker cites research by Gallup showing that the quality of middle managers determines 70% of the variance between high-performing and low-performing companies. Because managers can instill (or at least heighten) a sense of purpose and meaning in employees’ work, they drive the critical measure of employee engagement, which Gallup defines as a belief among employees that they’re doing meaningful work in a climate that supports personal growth. And that in turn leads to lower turnover, higher productivity and better profits.

This isn’t really an earth-shattering insight (even if the high level of correlation is eye-opening). We’ve all heard the maxim that workers don’t leave companies; they leave their managers. So this makes sense.

Now, it’s not always easy to connect an organization’s product or service to something “meaningful.” Not everyone works in an organization singularly focused on eradicating river blindness or rescuing otters. It’s a lot harder to talk about meaning when you work in a quarry or a factory making low-density polyethylene squeeze bottles for ketchup. And there are plenty of days when the work you’re doing—no matter where you're employed—isn’t going to feel particularly meaningful (notwithstanding the apocryphal story about JFK and the janitor at NASA who proudly explained that he was helping put a man on the moon).

But what’s most dangerous in Walker’s prescription is that companies simply need to hire (or train) better managers, and everything will be right with the world. He places the burden for increasing engagement on the individual manager.

That seems risky to me. There’s so much variability in people and environments that a “good” manager in one situation might not be a good manager in another.

What if instead we created mechanisms to increase employee engagement? What if we set up policies and processes so that it’s easy for people to grow in their jobs?

That’s the beauty of focusing on improvement. Every job does provide an opportunity to grow. Challenging people to improve the process by which they do their jobs harnesses their creativity and their innate desire to succeed, leading to higher engagement.

Unfortunately, too many organizations pay lip service to the notion of improvement without creating mechanisms to make it happen. They won’t spend the money to give workers enough training: A woman I know in the OpEx department of a $30 billion company needed three weeks to get approval for the $900 to attend a recent conference. Or they rely on suggestion boxes (“Where good ideas go to die”) rather than openly posted improvement boards. Or they don’t provide workers with the slack time they need to do improvement work—workers are paid for a 40-hour week, and it requires the papal seal of approval to get overtime pay for improvement work.

Cambridge Engineering in St. Louis is an example of a company that has created these mechanisms. Workers get training in lean and problem solving, of course, but even more importantly, they have 30 minutes every day to do “lean and clean”—work on improvement projects, or just clean/maintain/5S their areas. During their busy season, the company unquestioningly pays overtime to ensure that everyone has the ability to do their half hour of daily improvement.

Of course, these mechanisms don’t guarantee a high level of employee engagement. But they reduce the reliance on the talents of individual managers and increase the likelihood that employees will feel their jobs provide them with the opportunity for personal growth.

Toyota says that there are two aspects to every job: doing the work, and improving the work. Instead of relying on individual managers for employee engagement, let’s rely on this second component for higher employee engagement—and higher company profits.



Apollo 11: A Symphony of Work, a Ballet of Knowledge

I’m heading to Houston tomorrow for the LEI Summit—always an exceptionally inspiring and educational event. I’m particularly excited to visit Houston because I just saw the new documentary, Apollo 11. It’s amazing. Composed of video taken by NASA and the Apollo 11 astronauts, along with Walter Chronkite’s actual description of the events, the movie provides a kind of fly-on-the-wall perspective of the now 50-year-old mission. It’s riveting. 

My favorite part of the movie (aside from, you know, the actual landing on the moon) is the launch status check just before the Saturn V rocket lifts off. The “go/no go poll” was dramatized in the movie Apollo 13, but not by much—it’s pretty close to the real thing. (Here’s the status check for the launch of the space shuttle Discovery.)

You seldom get to see such tight coordination of knowledge workers. In physical assembly lines or manufacturing facilities, sure. But knowledge workers are more disconnected—communication and information flows asynchronously through emails and memos. Much of their work is done on a “push” basis, rather than pulled at the moment of need by customers, which reduces not just the drama, but also the inherent tension of the need to do the job precisely right. 

But in the go/no go poll, you see dozens (hundreds?) of knowledge workers coordinating their efforts as precisely as any high-tech robotic fabrication machinery. Each person (and each team) delivers exactly what the flight director needs exactly when he needs it. It’s a symphony of work, a ballet of knowledge, a miracle of coordination. 

And when you’re choking on yet another stupid “reply all” email, it’s utterly inspiring.



Lean is *Not* a Training and Development Activity

A prospect told me recently that he wanted to work with me to bring lean/continuous improvement to his company, but first he needed to “integrate it into his Training and Development” plans.

I told him that he was making a mistake, and that he was likely to fail.

If you frame lean/CI as a training and development activity, you won’t get the of buy-in and commitment from staff that you need. Lean becomes something nice to do, not something that must be done for the long-term success—and survival—of the company.

To be sure, when lean is done well, it IS a skills and human development activity. John Shook and others have written extensively about how lean is a socio-technical system, not just a collection of tools. And yes, some training is required to learn both those tools and the fundamental way of thinking.

But try getting a plant manager, or the leadership team of a company, to commit the necessary time and attention when they see lean as simply another training and development offering of the HR department. You’d have as much success telling your 10 year old that meditation and yoga are important for success in a PE game of kickball.

Whether or not you buy into the idea of lean as strategy, the results over the past 75 years show that it’s an unequalled tool for improving both the performance of an organization and the people who work in it. It’s not just a training and development activity. It must be integral to the way the company operates. You need to think of it as the way that you do business, period



How to Lay Off the Cognitive Donuts

Maggie Jackson, a journalist focusing on the effects of technology on the nature of our humanity, and author of the book Distracted, recently highlighted research showing that the mere presence of a cell phone—even if it’s turned off—lowers “fluid intelligence.” That is, the phone essentially siphons our attention away from what’s in front of us, making it more difficult to solve unfamiliar problems. 

She cites other research demonstrating that instantaneous access to information (I’m looking at you, Google) makes it less likely that we’ll struggle to solve a new problem. Even a brief online search for information reduces our willingness to engage in the cognitively challenging process of deep thinking. 

Jackson sums up this situation as follows:

 In our current culture, “knowing” is becoming something brief, perfunctory, neat, packaged, and easily accessible. Yet complex murky problems demand firstly the willingness not to know, to understand that the time for ease in thinking has ended and the real work of reflective cognition must begin.

And second, difficult problems demand tenacity, a willingness to struggle and connect and reflect on the problem and its possible solutions and move beyond the first answer that springs to mind. This is when we must extricate ourselves from automaticity in thinking and call consciously upon the side of ourselves that can decouple from tried-and-true answers, gather more information, test possibilities, and build new understanding.

Given this situation, is it any wonder that we have difficulty solving the complex problems that afflict our organizations? Is it any surprise that we struggle to avoid easy, reflexive “solutions” that seldom address the root cause of a problem? Who wants to take the time to work on an A3, or DMAIC, or 8D analysis when you can just do somethingand get plaudits from the CEO for being a take-charge employee who gets stuff done? Why chew through a (real) bagel when you can have a donut?

A few years ago, Karen Martin wrote about an experiment she ran with her workshop clients. Instead of teaching them a full suite of root-cause analysis tools, she’d do the following:

  1. Ask the participants to list a few problems and solutions to them. 

  2. Introduce the concept of PDSA (but without teaching any root cause analysis tools).

  3. Have the participants list the same problems as before, but this time with possible root causes—and suggest potential countermeasures to those root causes.

Each time, the participants come up with different countermeasures. Karen points out that even though their solutions may not be “best,” by simply inserting an additional step she was able to shortcut reflexive knee-jerk responses, and push people towards deeper, reflective consideration. 

Synthesizing these two articles leads me to the following conclusions:

  • People naturally gravitate towards simplistic answers. (This is the Type 1 vs. Type 2 thinking dichotomy that Kahneman and Tversky described.)

  • Today’s omnipresent, always-on, completely seductive technology makes it even more difficult to think deeply about problems. It distracts us, and it weakens our mental muscles. 

  • It’s incumbent upon lean thinkers to help people overcome the natural and the technological hurdles to deep thought. Karen’s experiment is one way to do this. I’ve suggested another approach. Neither of our techniques is the ultimate method of problem solving, of course, but they’re both useful in developing the right mindset. The skill set can follow. 



Why Toyota Kata May Be the Right Approach for You

One of the great benefits of the 2 Second Lean approach to lean is the way that it gets everyone engaged in kaizen with simple improvements. The genius of Paul Akers’ approach is the low barrier to entry for workers.

But as I’ve written about before, the problem with 2 Second Lean is the high barrier to entry for leadership. If leaders aren’t completely, continuously, and passionately involved as improvers and cheerleaders, it just doesn’t work. It becomes a trivial exercise distracting people from solving real problems. As one of my clients said to me, “Frankly, it’s hard to see how moving the garbage can closer is going to move the needle on our business.” Ouch.

And yet I’m also not a big fan of kaizen events. Sure, they have their benefits, but they tend to make continuous improvement discontinuous. Besides, copying an approach that was specifically designed to make life easier for Japanese consultants who flew back and forth between the US and Japan is like buying a buggy whip for your car, because it used to be a horse.

That’s why the presentations and the experiential workshops at the Katacon Summit last week in Savannah were inspiring. They reinforced the way that Toyota kata provides a path with a low barrier to entry by encouraging small experiments on individual obstacles. At the same time, kata ensures that leadership and internal skeptics can’t dismiss the improvement work as trivial by using the challenge to align the work with important business objectives. And finally, the requirement to have both a coach and a learner dealing with the gap between the current condition and the target condition means that employee engagement and managerial support will be high.

There’s no easy or universal path to creating a culture of continuous improvement. It’s a long, difficult trip, and each organization needs to find a route that works for its own idiosyncratic needs. My experience at the Kata Summit last week, though, makes me think that this is a great approach to try.



Lessons from Detroit: Three (More) Reasons Why Your Lean Efforts Fail

In 2014, The Greening of Detroit (TGD), an environmental non-profit, was pushing hard to reforest the city after years of neglect. To their surprise, the tree planters faced stiff resistance—about 25% of the 7,500 homeowners they approached rejected the opportunity to have a free tree planted in front of their houses. 

Residents weren’t stupid—they understood that the trees provided more shade, better air quality, and increased property values—but they still elected not to have trees planted for them.

As Brentin Mock explains in his CityLab article, the well-meaning volunteers made several fundamental errors that explain why their efforts failed. And that explain why so many of our well-meaning lean improvements fail as well. 

1. The Arrogance of the Enlightened: the volunteers from TGD “presumed to know what’s best” for the poor communities. The trees are unquestionably good for those families, but no one was included in the decision-making and planning processes. Instead, the volunteers simply assumed that their noble efforts would be valued and welcomed. They decided unilaterally what kind of trees to plant, which neighborhoods to plant in, and what the maintenance protocols would be without any input from residents.

How often do executives, internal CI leaders, or external consultants make the same mistake? It’s true that lean does make work easier and safer, but we often don’t include the workers in the initial discussions about what lean is, why lean is important, why the organization has chosen this approach, and where the initial efforts will be made. Sure, workers are introduced to lean before changes are implemented, but by that time the decision to go down that road has already been made in a conference room with executives and lengthy Powerpoint presentations. It’s no surprise when they “resist change”—just like the Detroit residents rejected the trees. 

2. Ignorance of Context: Black and brown residents remember when the Detroit municipal government chopped down trees and flew helicopters over their neighborhoods following the race rebellion in 1967. They believed that the city was doing this to make it easier to surveil their “dangerous” communities. The city was actually trying to stop the spread of Dutch elm disease by spraying them with DDT and then cutting down the dead trees. Consequently, when TGD showed up to plant new trees, people were skeptical. As the author explains, “it’s not that they didn’t trust the trees; they didn’t trust the city.” 

Whether it’s the outside consultants or a new leadership team, the promoters of lean are often ignorant of the context of continuous improvement efforts in the organization. Perhaps people were laid off years earlier during an effort to get “lean and mean.” Or maybe continuous improvement was used to justify making people work harder and faster, with fewer resources. Regardless of how often the new management or consultant spouts the “respect for people” mantra, workers will be suspicious that it’s the same thing as before. 

3. Neglecting the Externalities: Planting trees is not a simple, one-time activity. Residents knew that they’d be responsible for the long-term care of the trees, such as watering and fertilizing them when they’re young, raking up leaves in the fall, dealing with sidewalk damage from spreading roots, and clearing fallen limbs after storms. Not all residents wanted that responsibility—or, at the very least, they wanted a voice in the decision-making process, since they’d be taking care of the trees. 

Too often we neglect the externalities of a lean transformation. It requires that workers either do their jobs faster so that they have time for improvement activities during the day, or that they stay late to do their improvement work. But it’s not fair to workers to ask them to take on these burdens without additional support. Not many companies follow the lead of Cambridge Engineering, which pays people overtime to do their improvement work if it can’t be done during their regular shifts. Not many companies give workers time off to develop public speaking skills or fundamental MS Office competence, even though those skills are useful in leading shop floor improvements. 

It can be baffling why employees don’t embrace lean. By all standard measurements, it makes work easier, safer, and more fulfilling for workers. But if people won’t accept free trees in their neighborhoods, it’s no surprise that getting them to accept lean isn’t easy. Yet Detroit’s experience with reforestation teaches us three valuable lessons:

  1. Bring workers into the earliest discussions about lean. Don’t wait until the leadership team has decided to do it. Involve them in the decision.

  2. Learn the history of past improvement efforts. Are certain words toxic, or loaded with hidden meaning? Did certain production lines, departments, or plants have particularly painful experiences in the past? Were employees hurt by earlier improvement efforts? This knowledge will help you anticipate and mitigate problems. 

  3. Consider the new burdens lean places on workers, not just the benefits it confers. You’ll need to provide them with more support so that they can fully engage in improvement. Extra skills training, overtime pay, early morning or late night transportation to and from work, etc. are important signals that you understand—and support—the commitment they’re making to the lean effort. 



The Perils of Internal Disruption (Part 5)

“Disruption” has become another business buzzword that obviates the need for prudent, careful thought and consideration. If something is “disruptive,” then it must by definition be good. But when it comes to internal operations at least, disruption is often both bad for business and for employees, because it causes unevenness in work.

Last week, I wrote about how kaizen events can disrupt daily operations and overburden employees. In addition, they tend to signal that “continuous improvement” is actually discontinuous. (“We have a kaizen event this week. Next week it’s back to business as usual.”) I pointed out how companies can make kaizen a daily practice by setting aside standard time, or by using the Toyota kata approach of small experiments.

A final cause of self-inflicted disruption is management’s overreaction to noise in the data it measures. Managers capture all kinds of metrics, from the number of patient falls in a hospital ward, to the first pass yield in a production line, to the number of hits on a website, to the time it takes to repair a bicycle. They cover walls with graphs, and launch investigations when a number turns red or a trend turns downwards. However, not every change is meaningful. Too often, leaders react to every up and down in the metrics, asking for explanations and root causes that don’t actually exist. This kind of overreaction disrupts the organization and leads to activity that is more busy than useful. 

Some changes in metrics are just noise in an otherwise stable system. In his book Measures of Success, Mark Graban makes a compelling argument for more use of Process Behavior Charts (PBC), rather than bowling charts, bar graphs, or a table of numbers. PBCs (also known as process control charts) provide a holistic view of a system’s performance over time, allowing us to hear the “voice of the process.” This context enables management and front line workers to determine whether a change is significant, indicating that something has fundamentally shifted in the system and is worth investigating. As Graban writes, using PBCs subtly shifts the question from “What went wrong last week?” to “What was different last week?” Or even better, “How can we improve the system and its typical performance?” The result is less disruptive overreaction and empty explanations (what Professor Don Wheeler calls “writing fiction”) and more time spent on true value creating work.  

One of my own clients learned this lesson in tracking throughput at its repair facility. For months, the chart turned red or green based on the number of completed repairs each day, leading the VP of the facility to pressure workers, or congratulate them, based on the color in the bowling chart. This was both stressful and frustrating for his mechanics, since they felt as though they were working equally hard everyday.

Only after we made a PBC could we see that the the variation was simply noise in the system—there was no statistical significance to the change in the daily number of repairs. And with the chart, we were able to identify—and celebrate—when our changes truly improved the performance of the system.

Companies that create truly valuable disruptive products and services rightly reap outsize economic rewards. However, the headlong pursuit of external market disruption can blind leaders to the existence, and the cost, of internal disruptions caused by their own business practices. To be sure, some internal disruptions can be beneficial to the company by significantly streamlining processes. But when the disruptions lead to excessive unevenness in daily operations, they create distortions that stress employees, systems, and supply chain networks. By all means, pursue disruption for competitive advantage—but be careful not to disrupt yourself.

Part 1: Management By Walking Around
Part 2: Incentives & Discounts
Part 3: Batch Processing
Part 4: Kaizen Events



The Perils of Internal Disruption (Part 4)

“Disruption” has become another business buzzword that obviates the need for prudent, careful thought and consideration. If something is “disruptive,” then it must by definition be good. But when it comes to internal operations at least, disruption is often both bad for business and for employees, because it causes unevenness in work.

Last week, I wrote about how batch processing—whether in manufacturing processes, office processes, or even team management—creates unevenness and disrupts the smooth flow of work and value to the customer. This week I’d like to address the disruption caused by kaizen events.

What could be wrong with kaizen events? What company wouldn’t want to reap the benefits of large improvements in processes? Yet, kaizen events—in which employees in a given area stop their regular work for a full week in order to improve a given process—are the epitome of disruption. Kaizen events were invented by the original Japanese consultants who came from Japan to work with US companies in the late 1980s. It made no sense for consultants traveling all the way from Japan to the United States to work with a company for just a half-day or a full day. Instead, the consultant stayed for a full five days. To be sure, the benefits are real (if often unsustained). But kaizen events overload people in the week or two before the event by requiring them to produce extra in compensation for the upcoming downtime. Ironically, a weeklong event implicitly sends the signal that “kaizen,” which means “continuous improvement” in Japanese, is actually discontinuous. (“We’re doing improvement this week. Next week it’s back to business as usual.”) 

Companies that realize the greatest benefits from lean don’t do kaizen events (and that includes Toyota). Rather, they make kaizen a daily activity. Cambridge Engineering, an HVAC manufacturer near St. Louis, for example, has explicitly carved out 30 minutes everyday for employees to do “lean and clean.” Other companies are less structured about it, but still benefit from embedding improvement efforts into the fabric of daily activities without disrupting the overall flow of work. 

In fact, one of the great benefits of the Toyota Kata approach is that the size of each experiment is small by design. Crafting small experiments to move towards a target condition that can be achieved in a couple of weeks means significantly less disruption to the people and the processes in the organization. To be sure, sometimes implementing an improvement will necessitate shutting down a production line or a department for a few days—but not nearly as often as you might think.

Next week: Reacting to Noise in the Data.



The Perils of Internal Disruption (Part 3)

“Disruption” has become another business buzzword that obviates the need for prudent, careful thought and consideration. If something is “disruptive,” then it must by definition be good. But when it comes to internal operations at least, disruption is often both bad for business and for employees, because it causes unevenness in work.

Last week, I wrote about how sales incentives cause salespeople to stuff the company’s distribution channels with inventory far in excess of consumer demand. Volume discounts have the same effect, by encouraging customers to order more product than they need in order to get a larger discount. Not only do incentives wreak havoc on the supply chain through the “bullwhip effect,” the unevenness stresses employees and creates waste in customer service, distribution, and other functional areas.

This week, I’d like to address batch processing. It may be counterintuitive, but long production runs and large batches create disruptions in the flow of work compared to one-piece flow or small batch sizes. During the batch there’s little disruption, of course. But at the changeover, everything and everyone stops to move machines, change out dies, put different raw materials in place, etc. And it’s not just an issue in manufacturing—large batches create disruption in office and administrative processes as well. Shutting down a warehouse for two days to do physical inventory, for example, is incredibly disruptive, with ripple effects throughout the business, from supplier to customer. Similarly, most finance departments in large companies cut their activities to a bare minimum during the month-end close of the books, which often can take up well over a week.  

Toyota, most notably, has demonstrated the financial and quality benefits of one-piece flow over large batch processing in manufacturing. But working in smaller batches and avoiding disruption in office processes yields significant benefits as well. For example, many distribution centers use cycle counting to manage their inventory, avoiding the need to shut down the facility. Boeing’s finance department processes some of their financial information on a daily basis, rather than waiting to process a large batch at the end of the month. They look at what shipped each day, what materials were received every single day, and what bills were paid every single day. As a result of this (and other) changes, they reduced the time required to close the quarterly books from nearly one month to five days.

In HR, too, many companies are getting rid of the annual performance review in favor of shorter, more frequent discussions as often as once per month. This not only provides more timely and effective feedback for employees, but it eliminates the massive time commitment imposed on managers in November and December.

Finally, consider the typical weekly or bi-weekly meeting for a team: the manager is working with a one- or two-week batch of information. Not only is it difficult for the manager to process that much info in a single meeting, for everyone else, it’s about as interesting as reading last week’s newspaper. And by the time the manager gets the update, the information is at least a few days out of date. Far better to use daily huddles and stand-up meetings to exchange information.

Batch processing is rampant in most organizations and most functions. The sooner you can transition to one-piece flow (or at least small batches), the easier it will be for everyone to do their jobs.

Next week: Kaizen Events.



The Perils of Internal Disruption (Part 2)

“Disruption” has become another business buzzword that obviates the need for prudent, careful thought and consideration. If something is “disruptive,” then it must by definition be good. But when it comes to internal operations at least, disruption is often both bad for business and for employees, because it causes unevenness in work.

Last week, I wrote about how the unpredictability of “management by walking around” disrupts daily operations in an organization and leads to mura (unevenness), muri (overburden), and muda (waste). This week I’d like to address sales incentives and volume discounts. Incentives and discounts create tremendous disruption in a company’s business by distorting both incoming market signals and outgoing orders to suppliers.

Sales incentives—for example, bonuses to meet monthly or quarterly revenue goals—cause salespeople to stuff the company’s distribution channels with inventory far in excess of consumer demand. Volume discounts have the same effect, by encouraging customers to order more product than they need in order to get a larger discount. Both these practices wreak havoc on the supply chain through the “bullwhip effect.” Hau Lee, professor at the Stanford Graduate School of Business, illustrates this problem with a story about Volvo: in the mid-1990s, the Swedish car manufacturer found itself with excess inventory of green cars. The sales and marketing departments began offering special deals to clear out the inventory, but no one told the manufacturing department about the promotions. It read the increased sales as a sign that consumers had started to like green cars, and ramped up production.

The former president of Wiremold, Art Byrne, explains in his book The Lean Turnaround that he eliminated volume discounts and incentives for sales to book the largest possible orders. Instead, he pushed his sales team and his customers to provide a steady flow of small orders that would smooth demand and reduce disruptions. Large customers received cash rebates at the end of the year as a reward for their business, but without the supply-distorting incentives for large individual orders. This clever approach allowed Byrne to square the circle: provide incentives without increasing demand variation.

It’s worth mentioning that not all incentives are bad—only poorly designed ones are. In highly seasonal businesses, they can actually be a useful tool for leveling demand and reducing unevenness. I consulted to an outdoor goods company whose warranty department got into trouble every summer. Usage for this company’s products are overwhelmingly concentrated in summer months, and as you might expect, consumers wouldn’t send their products in for warrantee repair till the summer started. The sudden spike in warranty requests meant that lead time exploded from less than a week to five or six weeks. The company reduced the fluctuation in demand by sending out regular emails to customers from January-March, reminding them to send in their products for repair early, and promised a free logo t-shirt for people who sent in their item before April 1. This incentive not only increased the number of items that came in early, it got more brand logos on the street. With more level demand, the warranty department was able to keep its turnaround time to under a week, even during the peak summer season. 

Next week: Batch Processing.



The Perils of Internal Disruption (Part 1)

“Disruption” has become another business buzzword that obviates the need for prudent, careful thought and consideration. If something is “disruptive,” then it must by definition be good. But when it comes to internal operations at least, disruption is often both bad for business and for employees, because it causes unevenness in work.

Taiichi Ohno, one of the fathers of the Toyota production system, described three manufacturing evils that companies should avoid: mura (unevenness), muri (overburden), and muda (waste). In fact, unevennness is often a cause of overburden, which leads to the much of the waste that companies are so eager to eliminate. Unevenness in any aspect of a business—customer demand, process time, quality of raw materials, staffing, etc.—results in overburdening some resources at the expense of others, or alternating between overburdening and underutilizing a resource over time. For example, the spike in toy demand at Christmas puts enormous pressure on factories, warehouses, and logistics providers, to say nothing of front-line retail staff. Similarly, unevenness in machine availability will cause workers and machines in downstream processes to be alternately starved and overburdened with work. 

Unfortunately, while pursuing industry disruption for a better share in the market, leaders are often oblivious to the disruptions that their personal actions and business practices create for their organizations. These disruptions generate enormous unevenness for employees and processes, and make it more difficult for the organization to excel.

In this five part series, I’ll describe five types of disruption that can undermine a company’s performance.

Management By Walking Around
in their seminal book In Search of Excellence, Tom Peters and Bob Waterman popularized the concept of “management by walking around” (MBWA), encouraging leaders to get out of their offices and randomly walk around the company to see firsthand what’s going on. They specifically advise managers to make their walks unpredictable, both in terms of where they go and when they go. Peters and Waterman believe that if frontline workers are expecting management’s visit, they won’t see what’s really happening on a regular basis. They argue that frontline staff will work differently; they’ll clean up their work area; they’ll cover up small problems. Leaders won’t get an accurate picture of how the processes are operating. 

But this kind of unpredictability is a powerful form of disruption for the worker. If a senior leader randomly shows up, the workers will inevitably be anxious and stressed. They’ll work differently under the watchful eye of the boss, possibly creating variability in quality of their work. Or worse—they’ll stop working while they answer questions, affecting the timing of a production line, and creating unevenness for downstream workers. 

In contrast, organizations that have embraced lean thinking, like JD Machine, Stanford Medical Center, Lantech, and others, substitute standardized “gemba walks” for random MBWA. At these places, the leadership team has a regularly scheduled walk through the various departments to see first hand what’s happening. There are no surprises for the staff—they know who’s coming and when, with the result that these visits are smoothly integrated into daily work without disruption. Moreover, it’s both helpful and rewarding for front line staff to know that they’ll get to talk with the CEO or VP of Operations on a regular basis. 

Next week: sales incentives and volume discounts.