Mike Stoecklein recalls a story secondhand. It was, he thinks, at one of the last four-day seminars W. Edwards Deming ever gave. Deming was frail by then. Someone later told Mike that when the seminar got to the Red Bead Experiment, Deming stood up slowly in front of the executives in the room and said something like:
“Young Jerry Stoecklein, nine years old, he understands the lessons of the red beads. Why can't you?”
Mike has no video, no audio. Just the account. But the piece he does have firsthand is the paper trail. A few years earlier, Mike had been using Deming's red bead experiment to homeschool his nine-year-old son, Jerry. Jerry and Mike then co-wrote a paper for the American Association of School Administrators, titled “What Do Grades Mean?” Jerry, then in fifth grade, handwrote a letter to Deming asking permission to quote him. Deming wrote back. Granted permission. Apparently held on to the boy's name.
That's the scale of Deming's teaching. A nine-year-old. A handwritten letter. A ballroom of senior people being told they were being outperformed by a fifth-grader.
How the lines actually run
I've been doing this podcast for twenty years. It started in July 2006 with Norm Bodek. What has surprised me, sitting with the archive, is how many of the people I've talked with across those two decades came to their understanding of the work through Deming, or through someone Deming had taught, or through someone who had taught someone Deming had taught. The lines run through everything. The Lean movement has never been only about Toyota. Toyota itself will tell you that. What Deming gave us, or tried to give us, against a steady American headwind, was a theory of management that was supposed to sit under all of it, and that most American executives found unwelcome then and still find unwelcome now.
This post is part of a series marking twenty years of these conversations. I'm writing it because a feature like this has to reckon with what the long arc actually shows, and what the long arc actually shows, in Deming's case, is that the shadow keeps moving.
The basement office in Washington
Kevin Cahill is the closest line. Deming was his grandfather.
Kevin has led the Deming Institute as Executive Director since 2011. As of late 2025, he is transitioning that role to a successor while remaining President of the Board. When we talked in January of 2016, he described something I've kept thinking about. As a kid, he didn't know what his grandfather did. He knew there was a small office in the basement, and a large chalkboard covered in mathematical figures. Kevin and his brother both decided, very early, that whatever it was their grandfather did looked like one of the most boring things a person could choose to do with a life. That assessment survived until Kevin was well into his own career and started reading his grandfather's books.
The books had been written in the basement. So had most of the seminars, the consulting decks, the letters that traveled back to people like Jerry Stoecklein. It took Kevin decades to understand what he had been in proximity to the whole time.
There's something in the domestic scale of that which matters for understanding why Deming has been so hard to translate into American management practice. The basement office was not a stage. The man was not performing. He was doing a specific kind of careful, mathematical work with moral consequences, in a house, for forty years. Most American management systems don't survive that kind of coherence. They require the person to be a different kind of person at work than at home.
Madison, 1986
Mike Stoecklein came to Deming in Madison, Wisconsin, in 1986. He had just moved from California, where he had been a medical technologist in the Navy. He was starting to move into management, running into system problems his training hadn't prepared him for, and a professor at the graduate school who was a statistician brought him to hear Deming speak. Deming was in town to kick off the Madison Area Quality Improvement Network. Out of the Crisis had been published a few years earlier.
For the next nine years, until Deming died, Mike wrote him letters. Deming always wrote back.
He also started volunteering to help with the four-day seminars — which meant he spent real time in the same rooms, watching Deming work. The thing Mike absorbed was not the content of the 14 Points. It was the posture. The way Deming opened every seminar, large or small, with the same question:
“So why are we here?”
The answer, delivered after a long pause and sometimes after taking a few offered responses from the room, was this:
“I believe we're here to learn. But we're also here to have fun.”
Mike told me this almost thirty years after the fact, and you could hear, in how carefully he repeated the words, that he had kept them.
The other thing he kept was the way Deming listened. Every time Mike saw him, Deming was writing notes. Pulling them out of his suit coat pockets, writing, putting them back. “People thought they were learning from him,” Mike said. “But he was learning from them.” That image — the man in his late eighties, in front of a paying audience of executives, still taking notes — is a better gloss on “continue to learn every day” than any aphorism.
The red bead story is the one I want to come back to. Mike ran a home version of the experiment for Jerry and a few of Jerry's friends, the way Deming ran it for executives. He fired some of the children. He promoted others. Same script. Then Mike and Jerry wrote the paper, and Jerry wrote the letter, and Deming wrote back. And somewhere along the way Deming apparently decided that a fifth-grader in the Midwest who understood that grades are an attempt to assign individual responsibility for something the system controls was more clued in than the ballrooms he had been addressing for forty years.
He was probably right.
The documentary that kept working
Clare Crawford-Mason is the reason most Americans have ever heard of Deming at all.
She was the NBC senior producer behind the 1980 white-paper documentary If Japan Can, Why Can't We? — the one where Deming, in the last twelve minutes, appears on camera and explains why American industry is losing to the Japanese. NBC had never, before or since, had a documentary generate that much response. Thousands of phone calls, she told me. Requests for transcripts. Requests for Deming's contact information. Ford called. The Fortune 500, she said, “lined up outside his basement door” trying to figure out what to copy from the Japanese.
Deming was nearly eighty. Essentially unknown in his own country. A week after the documentary aired, he was consulting with the largest automakers in the world.
When we talked in March of 2013, Clare told me the story of how the documentary came to exist. NBC had asked her to produce an hour on Japanese productivity, under the working title “Whatever Happened to Good Old Yankee Ingenuity?” The assumption — the default American assumption in 1980 — was that the story was cultural. The Japanese were just different. Clare's crew went looking for Americans who might have an alternative explanation. The dean of the American University business school, Herbert Striner, suggested they talk to an elderly statistician who lived up the street from the university, a man who had worked with Japanese companies after the war. Clare visited with Deming for a total of twenty-three hours across the production. She reported back to NBC that no one in Washington knew who he was. She had asked the head of the President's Council of Economic Advisors about him; he thought Deming might be the son of a different Deming. The president of NBC said:
“Oldest story in the world. A prophet ignored in his own homeland. We'll put him in the documentary.”
Clare said something else that has stayed with me. Neither she nor the documentary's anchor, Lloyd Dobyns, actually understood what Deming had taught the Japanese. Not during the filming. Not for five more years. Only later, when she and her husband Bob began working with Deming on the Deming Library video series, did she start to understand that statistical process control was a small part of a much larger theory of management. She spent the next forty years trying to help more Americans understand it, and in our 2013 conversation she told me she thought the country still mostly hadn't.
The documentary is on YouTube now, which took Kevin Cahill years of negotiation with NBCUniversal to arrange. The tape keeps doing the job. And for decades after that first interview in 1979, Clare kept doing the job alongside it — the Deming Library video series with her husband Bob, the PBS documentary Good News: How Hospitals Heal Themselves, the book The Nun and the Bureaucrat, the speaking, and a follow-up book, The New Wisdom, that she was still writing into her eighties.
The executive who got told to leave
John Dyer was a process engineer at GE Major Appliances in Louisville in the mid-1980s, building refrigerators on second shift, when a massive automation project he was supposed to help run failed in the way massive automation projects tend to fail.
The GE corporate audit team that came in to figure out what had happened pulled him onto the team. They spent six months. The findings came down to three things.
First, if you have broken processes and you automate them, you have broken automated processes. You just make junk faster.
Second, the company had not involved the employees in the changes, and the employees, predictably, were rooting for the line to fail. John remembers one night, after he and the maintenance guys had spent hours getting a recalcitrant machine back online, finding a refrigerator stuck on the conveyor a hundred feet away, with a dozen workers standing near it, none of whom had bothered to give it the shove it needed. The system was not theirs. Why would they help it.
Third, and this is the one that stuck: a survey of the plant's engineers found that over ninety-nine percent of their time was spent firefighting. Less than one percent on anything that would actually fix anything.
That third finding earned John what amounted to a corporate dream job. GE sent him on the road for two years to collect best practices. He went to every division of GE. He spent time at Crotonville. He sat in on the original Six Sigma training at Motorola University, back before there were belts. And along the way he went to a Deming four-day seminar, and then to the smaller follow-on course, and came home with a photograph of himself and Deming from 1991 that Deming had inscribed by hand.
After that first four-day, John went back to Louisville and convened the top executives of Major Appliances in a room. He walked them through Deming's 14 Points one at a time. After each point, he polled the room on two questions: how important do you think this is, and how well do you think we do it.
Only one of the fourteen got a ten on importance. Over half got twos and threes. Moving away from “Management by Objectives” drew the loudest objections. John told me he has run the same exercise with many executive groups in the years since, and the scoring has barely moved.
When Jack Welch eventually wrote about Deming in one of his memoirs, he described Deming's work as “too theoretical.” John laughed about this when we talked. Of course that's what Welch said. That's what you say if you don't want to change.
“Knowledge without theory,” Deming had said in a seminar John attended, “is kind of a waste of time.”
Theory was also why, when John sat in another four-day seminar and watched an executive in an expensive suit walk up to the open microphone and ask, “Who are you to tell me how to run my company? What makes you think you know all of this stuff?” — Deming didn't try to make the man comfortable. He looked at him and said:
“Sir, you clearly do not get this. It is clear to me that you'll never get this. I'm wasting your time and you're wasting my time. I think you should leave.”
The man turned beet red. He stormed out. The room applauded. Every person in that room had a guy in their company they wished they could do that to.
That scene, John told me, more than anything else, is what he took home.
Every word chosen on purpose
John saw Deming run the Red Bead Experiment twice. The second time, he took detailed notes, because he had started to realize that every word Deming used was chosen on purpose.
The experiment is famous enough that I'll compress it. Six willing workers. Two inspectors. A paddle. A container of white beads and red beads, eighty percent white, twenty percent red. The workers dip the paddle in, pull out fifty beads, and take them to the inspectors. The inspectors count reds. The workers are praised or disciplined or fired based on their results. None of it has anything to do with the workers. The system controls the outcome. The only real variation is whether the workers happen to pull more red beads on any given run or fewer.
What struck John on the second viewing was the scripting. Willing workers. Minimum education required. Agitate the paddle and then do nothing. Do nothing. The paddle must come out at a forty-four degree angle, eight centimeters above the container. Gravity is cheap. Deming was both running a statistical demonstration and constructing, in real time, a parody of the kind of over-specified, procedure-obsessed management that was causing the very problem the experiment was about to prove couldn't be fixed by tighter procedures.
And when the executives in the room finally got it, when they saw the control chart go up at the end and understood that what they had been watching was common-cause variation, that the worker they had been about to reward and the worker they had been about to fire were interchangeable, Deming had his point.
Don't blame the worker. Look at the system. That is the sentence version. But sentence versions don't do the work the experiment does.
I have run the red bead experiment myself dozens of times in my workshops. What I have learned is that the number on the control chart is almost beside the point. The thing that sticks is the feeling. The willing worker on stage who tried to follow the procedure and felt terrible when she pulled thirteen red beads and got put on probation. That feeling is what most hospital staff I have worked with have lived inside their entire careers. They tried. They followed the procedure. They got rewarded or punished by their red bead counts. It had nothing to do with them. No one told them.
Drive out fear, as a clinical diagnosis
Robert Maurer worked as a clinical psychologist at Santa Monica-UCLA Medical Center, where he directs behavioral sciences for the Family Practice Residency Program. He attended Deming's four-day seminars near the end of Deming's life, though he didn't have much opportunity to speak with him directly. What he has spent decades studying, through his own clinical work and through prospective research on people who thrive across long stretches of their lives, is the neurobiology of fear and change.
Which makes him, by the long route, a translator of Deming's eighth point.
Deming said: drive out fear, so that everyone may work effectively for the company. Maurer's work fills in the mechanism. When people are afraid, the parts of the brain that can learn shut down. The amygdala takes over. You cannot be simultaneously scared of being fired and curious about the process. The organism cannot hold both states at once. Most American management systems are built on the premise that fear is a useful motivator. The evidence from neurobiology is that fear is the single most effective tool for making sure the thing you want to happen does not.
Deming understood this without the neurobiology. He just watched people. He watched what happened in a factory when you fired one worker in front of the others. He watched what happened in a seminar when he told a CEO to leave. The others did not then perform better. They went quiet. They went careful. They started hiding things. He put it in the book in 1982. American business picked up most of the book and ignored that chapter. The companies kept the part about quality and threw out the part about how to make quality possible. Then they ordered pizza for the team.
Safety as the canary
Mike Taubitz spent over forty-three years at General Motors, starting in engineering and ending as Global Safety Director. When we talked in May of 2013, what he wanted me to understand was the specific moment GM started taking safety seriously, and who made that happen.
Paul O'Neill, at that point chairman of Alcoa and a member of the GM board, started asking GM's top management pointed questions about workplace safety. Mike, who had been in corporate safety for years, used O'Neill's visits to finally make the argument that safety could not be a staff function and a side metric. It had to be owned by operating management, by the people who actually ran the plants. Only when O'Neill's questioning came from a GM director's chair did the argument land.
The thing people miss about O'Neill's safety-first reputation at Alcoa, Mike said, was why. O'Neill understood, the way Deming understood, that if you cannot fix injuries to your workers, you cannot fix anything. Safety is the canary. If your company's management system is failing your people, the first place it shows up is in the injury rate, because that is the first place the cost appears in a form no one can hide. Fix the safety system and you are, necessarily, fixing the management system. You can't do the first without the second.
This is the thing American business keeps getting wrong about Deming. The 14 Points are not fourteen things on a menu. They describe one integrated system. You can't pick three. You can't skip “drive out fear” because it is uncomfortable and keep the control charts because they are interesting. The integration is the point. Taubitz watched GM take forty-three years to partially understand this, and his voice in 2013 was not one of triumph.
A real Deming company, for comparison
In February of 2007, fairly early in the life of this podcast, I interviewed Eric Christiansen, the president of a translation services company in Iowa called Omni Lingua. The company had been bought in 1994 by an owner who had worked and studied directly with Deming. The owner's reason for the purchase was specific. He wanted to run a Deming company, and he had concluded that the only way to do that was to own one.
Eric walked me through what it actually looked like. No annual reviews. No sales commissions. No production bonuses. Company-wide profit sharing, the same amount for everybody. Single-source suppliers, chosen on philosophy and retained through long-term relationships rather than yearly bid cycles. A sales team on salary, hired specifically for people who were tired of having their compensation plans tweaked every month. Service staff who were empowered to tell a client no when telling the client yes was going to hurt them downstream.
The hiring pushback Eric described was predictable. Were they just hiring salespeople who couldn't sell? What he found was the opposite. The experienced salespeople, the good ones, were exhausted by the monthly-quota-and-tweak cycle, and would take a pay structure that let them do the actual work in exchange for having the structure left alone.
Most Deming retrospectives treat the ideas as an influence, a sensibility, a tradition. Eric was describing a company that had simply done the thing. Eighteen years into the experiment, when we spoke, it was still running.
The price of admission, evidently, was buying the building.
The next generation
More than thirty years after Deming's death, most of the people carrying his thinking forward did not sit in his seminars. They read The New Economics in graduate school, or stumbled into Out of the Crisis through a boss, or found Deming through Don Wheeler's books on Process Behavior Charts, or were handed the “If Japan Can” tape on VHS by a mentor in the mid-1990s.
John Dues is the Chief Learning Officer of the United Schools Network, a four-school K-through-eight charter network in Columbus, Ohio. When we talked in February of 2022, he was using Process Behavior Charts to help teachers and administrators see what their student-performance data was actually telling them, as opposed to what the state accountability regime was pushing them to see. Test scores rise and fall. Most of that movement is common-cause variation. Treating every up-tick as a triumph and every down-tick as a crisis is one of the most effective ways I know to exhaust a teaching staff and chase improvement in circles. John is in his forties. He is not a silver-haired veteran of the four-day seminars. He is applying Deming's thinking to a field Deming never worked in, and it works, because the theory travels. He's also written Win-Win: W. Edwards Deming, the System of Profound Knowledge, and the Science of Improving Schools.
John Willis is the author of Deming's Journey to Profound Knowledge and a co-author of The DevOps Handbook. When we talked in March of 2025, what he described was a personal trajectory that became a movement. He was on an early DevOps Days panel in the 2000s with Gene Kim, who later handed him a reading list that included Ohno and Steven Spear. The more he dug, the more the trail led back to Deming. Over time, Willis told me, other DevOps people kept finding the same thing. Every problem they were trying to solve at the management-system layer for software that shipped hundreds of times a day, Deming had already described. Willis eventually wrote a book in a Michael Lewis style to try to explain why.
Marcia Daszko did know Deming personally. She attended twenty of his four-day seminars, co-founded the Bay Area Deming User Group, and has spent decades trying to get executives to abandon performance reviews, ranking systems, and management by objectives. When we talked in March of 2024, what she kept coming back to was the same distinction Deming made: accountability lives with leaders, because only leaders can change the system.
“If people don't want to be held accountable,” she told me, “they are not leaders. They are executives with titles and positions and lots of money.”
She is still at this work. It's a long road, but the length is the point.
Balaji Reddie founded the Deming Forum of India in 1999, after spending years studying under Henry Neave of the British Deming Association. When we talked in September of 2021, he pointed out something I found striking: outside Japan, India has the largest number of Deming Prize-winning companies in the world. And yet, Balaji said, many of those companies are so focused on the prize that they miss the philosophy. The 14 Points are not the whole of Deming. Neither is the statistical work. The real work is the System of Profound Knowledge, and that takes a lifetime to learn.
Skip Steward, the Chief Improvement Officer at Baptist Memorial Health Care in Tennessee, came to Deming through Don Wheeler's workshops and through the “If Japan Can” tape on VHS. When we talked in 2018, he told me about handing out my own book on Process Behavior Charts alongside Wheeler's Understanding Variation to hospital leaders who were looking at readmission rates and patient falls and convincing themselves they were improving, when what they were actually looking at was stable systems fluctuating around their means. Skip's framing of it, which I think is exactly right:
“The process is telling you: if you don't like me, then you better change me.”
Alfie Kohn, in a completely different part of the forest, has been making Deming's psychology argument for forty years. His books Punished by Rewards and No Contest are, from one angle, a working-out of what Deming meant by “drive out fear” and “abolish the annual rating” and “competition is the destruction of systems.” When I interviewed Alfie in January of 2009 about Chicago Public Schools paying students for good grades, he ended the episode with a dead-on Deming impression that was partly a joke and partly the most direct tribute any of my guests have paid the man. He was making Deming's point, in Deming's voice, about something Deming would have had plenty to say about.
The lines run through everything. They do not stop running just because the man himself is gone.
What the shadow is doing
The thing you notice, after twenty years of these conversations, is that Deming's shadow does not actually belong to Deming. It belongs to the people who keep pointing at it.
The seminar-goers are aging out. The number of people who can still tell you what it felt like to be in the room with him is dropping every year. Kevin Cahill is already transitioning the Institute's Executive Director role to a successor. The four-day seminar is something future generations will only ever know on YouTube or through the Deming Library. And yet the ideas keep picking up new carriers. A charter-school principal in Ohio. A DevOps author in Kentucky. A safety director who spent his career inside GM. A young translator's boss in Iowa who bought a company specifically so he could run it the right way. Most of them did not sit in the ballroom. All of them are doing the work.
What I find myself asking, twenty years into this conversation, is whether the American management system that Deming described in 1982 has actually changed, or whether the thing that has changed is only who is willing to say, in public, that it is broken. Deming said it every day of his working life. A handful of people listened. The rest of the country waited him out. Most of his 14 Points are still rejected in executive boardrooms. Performance reviews, ranking, and management by objectives are still the default. The fear he wanted driven out is still the primary motivator of most people's working weeks.
And yet the ideas keep finding carriers. A nine-year-old got them in 1993. A DevOps engineer picked them up in 2007. A school principal found them in 2020. Something is working, slowly, in the way Deming said it would have to. Long range. Requiring, as he put it, a great deal of patience.
Maybe that is the thing the shadow is doing. Waiting. Making itself easy to find when the next generation decides it has had enough of management as it is, and starts looking for something that actually describes how systems work.
The series continues next week with highlights from podcasts with Jeff Liker.
