The Parable of the Red Beads
The Parable of the Red Beads
A Bad System Will Beat a Good Person Every Time
In lectures, I have often made use of a simple experiment to demonstrate that is all too easy to blame workers for faults that belong to the system… Apparatus: Red and white wooden beads, in a box. Paddle: 50 bevelled depressions, 10 x 5. One scoop with the paddle pulls up 50 beads. Advertisement on blackboard or overhead projector, to draw volunteers from the audience:
Vacancies, 10. Applicants must be willing to work.
Educational Requirements minimal.- Dr. W.E. Deming. Out of the Crisis, 2nd ed. (p. 345)
We had in the Red Bead a horrible example of the worst procedures anyone could put together, done purposely. I made it that way purposely to make the point.
- Dr. W.E. Deming. Lessons of the Red Beads video. (23:42)
HOW many management thinkers can you think of that are so closely associated with a teaching simulation or game that they become almost synonymous? Mention the name Deming among those who have an inkling about the man and his books and you’re almost guaranteed to elicit an excited interjection about his famous Red Bead Experiment. It happened just the other day for me when I was chatting with a tech company CEO who couldn’t contain their enthusiasm describing Deming’s deadpan delivery of wry lines as the foreman.
A staple of Dr. Deming’s four day seminars, the experiment is like a managerial Penn & Teller show, misdirecting your attention toward the exercise of watching six hapless, “above-average, willing workers” trying to sift 50 white beads out of a bucket containing an 80:20 mix of white and red beads while teaching you over a dozen lessons on the deficiencies of the prevailing style of management. If you have never seen or participated in the experiment, take a few moments to watch Deming himself facilitating a run at one of his seminars:
Most people immediately pick up on the ruse that barricading and haranguing the workers (misdirection!) won’t help when the problem is in the bucket, and that in the real world, we don’t do these sorts of things to people. Well, yes and no. Per Deming’s quoted-thoughts above, the experiment is an assembly of the “worst procedures anyone could put together” to drive home the point that without profound knowledge management is flailing away very much like the comical Foreman who makes all kinds of cringe-worthy overtures to reduce defects.
What is the Foreman’s Theory of Management?
A go-to question I pose to managers I am teaching Deming’s theories to is to have them define or explain their theory of management, ie. the ideas or concepts they use to predict the outcomes of their actions or interventions. After viewing the video above, how would you characterize the theory of management of the Foreman (as portrayed by Dr. Deming) ?
What I see is an inversion of Deming’s proportion of troubles and possibilities for improvement, ie. 94% attributed to special causes by people, 6% system or management, along with several corollaries:
Extrinsic threats/rewards change behaviour positively and this improves quality;
Past performance is an indicator of future trends;
Conformance to standards over in situ improvement;
Management by Objective (MBO)
Management by Results (MBR)
Management by visible figures alone.
Lessons of the Red Bead Experiment
I’ve found no other demonstration communicates more for the time invested than the Red Bead Experiment, so much so that I try to use it very early on in a coaching engagement because it provides a vivid experience to reach back to when we dig deeper in our learning of Deming’s theory.
Here’s what you can expect to learn - perhaps not all at once, but over time - from participating in the experiment:
Systems which seem to be chaotic and unpredictable tend to be the opposite - you just haven’t had the tools to see it, yet.
The process has more to do with differences between people’s performance than any innate skill. The workers put in all they had to offer - they could do no better under the circumstances.
Rating and ranking demoralizes people for performance of a system; rate and rank the system, instead.
Pay for performance does not improve a system, it merely rewards luck while demoralizing people.
Past performance of an individual is not an indicator of future results when the system remains fixed.
Rigid procedures that prevent suggestions for improvement guarantee mediocrity and failure.
Managing by Results (MBR) by reacting to daily defect counts does not contribute to improving the system - it’s tampering.
Managing by Objective (MBO) (3 red beads or less!) without providing a method does not contribute to improving the system - it’s tampering.
Managing without knowledge of variation guarantees MBR and MBO, and thus tampering and mediocrity.
The difference between mechanical sampling (dipping the paddle) and randomness.
The pointlessness of inspecting quality-in with the two inspectors; quality was already pre-determined by the system/process.
The Foreman is himself a product of the host system, perpetuating the practices that contribute to stasis and poor quality.
Slogans, exhortations, chiding, cajoling aimed at people to improve performance does nothing but humiliate and shame them for circumstances beyond their control.
It’s very easy to destroy pride in work by ignoring system effects.
I’ve facilitated over a dozen runs and have had reactions from participants much like “Ann’s” whom Dr. Deming quoted during a Lessons of The Red Beads recap at one of his seminars. Despite the gulf of time and the fact my participants typically work in tech, the reactions have been almost identical:
"When I was a willing worker on the Red Bead [experiment], I learned more than statistical theory. I knew that the system would not allow me to meet the goal of three red beads, but I still felt that I could. I wished to, I tried so hard, I felt responsibility - others depended on me. I'm going to try and keep the place open, otherwise we'll all lose our jobs. My logic and emotions conflicted; I was frustrated. Logic said there was no way to succeed, emotions said I could by trying.
After it was over, I thought about my own work situation, how often our people were in a situation they cannot govern, but wish to do their best, and people do their best. And after a while, what happened to their drive, their care, their desire? For some, they become turned off, tuned out. Fortunately, there are many who only need the opportunity and methods to contribute with." Lessons of the Red Beads (22:02 - 23:42)
I’ve had participants confide the same feelings of shame and inadequacy, and sometimes outright dismay. In one instance while facilitating with a colleague at a conference in Ottawa we had one participant storm off the stage calling the whole thing a sham (well, they weren’t wrong…). When I ran a session with a team of nine senior managers and their director, it was incredibly impactful. They laughed at recognizing their own foibles and it served as a hook to get them deeper into the work of real system improvements.
Learning by Doing
Deming is said to have characterized his experiment as “stupidly simple”, and it is. However, it is also a sophisticated ruse that “tricks” participants into learning very profound things about our prevailing ways and means of management. Dr. Deming deliberately designed it to make the lessons painfully clear so real learning could commence.
How would the Red Bead experiment be received in your organization? What analogs to the red beads would be apparent? How many of the management practices Deming designed-in to the simulation are used today? Have the effects on employees been similar? How about management?
Extra Credit Material
Steve Prevette’s classic rendition of the Red Bead Experiment strongly influenced my first attempts at facilitation of the game. Be sure to check out all six parts, an artifact of when YouTube maxxed out at 15m of video!
Don Berwick’s 2016 reinterpretations (here, and here) used blue beads for white and threw in a twist with seven, and then nine red beads as the maximum acceptable defects.
A friend of a friend created a 3D printing model of the paddle Deming used in his demonstrations. All you need for your own kit is a mix of red and white beads and a container to hold them.
Don’t have a 3D printer but still want to run the experiment? I created a spinoff called Defect Dice that can be run using five regular dice and four participants. Feedback on how to improve the game is welcome!