Comparing what Eric Trist’s (Socio Technical Systems Theory) self-managing teams and Deming’s Red Bead game have in common – and how they relate to the early history of Agile!
I’ve always been surprised at what seem to be the parallel tracks in thinking between W. Edwards Deming and the socio technical systems thinking developed by Trist et al in the early 1950’s, namely their preliminary studies following the nationalization of the British Coal industry. I have always wondered why their work isn’t more closely connected.
Deming’s work in manufacturing
Deming’s work in the USA and then ultimately applied in ‘anger’ post World War II, helped deliver the revolution in Japanese manufacturing. The USA then rediscovered him and asked for his help in their transformation in the 1980’s. His focus was on helping industry understand their organizations as systems with a focus on understanding what customers wanted, delivering this to the highest quality possible then focusing on delighting them.
This is the foundation of lean systems thinking and now Agile. Yes, Agile is not new!
Fundamental to this work is the importance of the ‘production system’. This means understanding others’ contributions to the delivery of value to the customer and working together seamlessly to achieve this. Although developed to describe the manufacturing system, I believe “production system” translates to all organizational systems.
Meanwhile, Eric Trist was on the other side of the world in the UK studying the effects of work design on both system performance and on people. There are many parallels in their thinking, mainly about the importance of understanding the impact of fragmenting work design on people and performance, and of course the customers that they are there to serve!
It is not clear whether each was aware of the others’ work. Trist’s ultimate focus became human relations, whereas Deming’s was on all variables in organizations affecting performance. Deming’s focus includes how we treat people and how this impacts performance. However, the parallels between the two are enormous. I think they both reach very similar conclusions but from different angles.
Trist’s work in coal mines
Trist’s work observed that there was a significant difference in both performance, output, sick absence and socio-psychological wellbeing when the mines in the UK were nationalized and the work design changed significantly as a result of mechanization. Because of the functionality of the machines, the work of the miners changed to being single task focused. This reductionist approach to work design has no apparent need for a team to work together.
This was very different to how miners worked previously. Their previous work design enabled self-managing teams to plan and do the work together at the coal face when the mines were privately owned. The teams covered the whole task at the coal face and they were interdependent. All members of the team were able to carry out all tasks.
It is possibly a counter intuitive conclusion that machines would not improve efficiency of the whole system, but the work of Trist and colleagues proved that the removal of the self-managing team into functionalized work design created performance deficits and had a significant impact on the wellbeing of the miners and of the performance of the system. Production went down, staff absence went up and costs increased. It was the idea of managers, who were removed from understanding the impact of the design of the existing system on performance. It was a design they created based on assumption rather than knowledge. In the absence of knowledge, assumption is always the king! In my experience – this king ruled in business then and today.
Trist, Deming, and Agile
So, what has this to do with Deming and his Red Beads game and what has it all to do with Agile?
Deming conducted the role play experiment with participants in his 4-day seminars and it continues to be used to help people understand the flaws in traditional management thinking and the lack of employee or team self-direction in work design to this day. The red bead experiment goes a long way toward explaining the underlying basis of Deming’s methods.
The Red Beads Experiment
Imagine thousands of beads in a container representing incoming raw materials to a business:
- 80 percent of the beads are white, 20 percent are red
- The work goal calls for an employee to use a spatula-like paddle to scoop beads from the container and transfer them to another container. Each dip of the paddle represents one workday of a willing worker.
- The work standard calls for the company to ship 50 white beads every day. The buyer only wants white beads and will not accept or pay for red beads under any circumstances.
- In the experiment, various individuals role-play willing workers, inspectors, recorders and managers. With all the participants present, the willing workers are told to get to work.
Of course, each time a willing worker dips the paddle into the container, they inevitably end up with some red beads. As the game progresses, managers use all the standard methods of responding to the “lack of success” of the willing workers. They scold them for getting red beads, they offer incentives to encourage them to get all white beads, they clarify the goal of shipping only white beads, they discuss adding inspectors, they discuss conducting additional training, they develop detailed policies and procedures. The managers also threaten to fire the workers getting too many red beads, they praise the workers getting fewer red beads than their previous attempts or fewer than another willing worker, and so forth and so on.
In reality, each worker has little or nothing to do with the number of red beads that end up in their final product. The process produces random results. Even when workers try to do their best and are provided the incentives, training and motivation to do so, a flawed system will rarely, if ever, produce quality results. And well-meaning managers expend a tremendous amount of time, effort and energy on things that do not matter in terms of reaching production goals.
Even when workers try to do their best and are provided the incentives, training and motivation to do so, a flawed system will rarely, if ever, produce quality results.
Blame it on the process, not the people
Deming’s red bead experiment illustrates one of his most famous discoveries: bad people don’t cause most problems; Bad process and work design do!
The following is an example of a comment I’ve heard from people who have played the game, “I was frustrated that they would not let them do their work and they were powerless to change anything.” This is a very common experience. As Deming read from a “willing workers” letter”
“People wished to do their best. I thought about my own work situation, how often people are in a situation they cannot govern but wished to do their best, and people do their best. And after a while, what happens to their drive, their care, their desire? For some they become burned out, tuned out”.
The red bead experiment takes it to an extreme, but the point is being forced to work in a broken system that you are powerless to change.
Back to Trist and the impact of work design on the socio-psychological in people and its links to what we know about the functionalisation of work and the flaws in assumptions, as well as what the assumptions create in work design.
The experts on work are the people who do the work, if they are given the end to end responsibility to deliver value (self-managing teams). We should have the right measures in place (in their hands) to make sure the work is delivered to the highest standard possible. It’s workers’ responsibility to act in the system and leaders and managers responsibility to work on the system – removing barriers and problem solving that is outside the responsibility of people doing the work.
So, what has all of this got to do with the history of Agile?
The creation of agile processes, systems and organizations demands a different way of thinking about the design and management of work. It puts the responsibility of making decisions about the day to day work in the hands of people who do the work, with the measures they need in their hands to ensure decisions are based on knowledge, not assumption. It relies on team working, pulling people together with the skills, knowledge and experience to do the job from end to end. Agile also relies on being able to pull others into the team if the demands of the project or job or customer change.
The parallels between this and the work of Trist, even though it was in the middle of the last century, are clear. What is more astonishing is that we are still stuck to the traditional mechanistic paradigm of thinking about systems and work design. Was it the advent of mechanization way back then that created an adherence to a mechanistic theory of organization and still reigns today?
Trist’s work in coal mines (and other industries) points to less of a mechanistic theory and more of an organic theory of organization. That was in the 1950s! Close to a century later, it is still the dominant paradigm, despite the efforts of Deming and others.
Let’s hope that its modern iteration, Agile, succeeds where others have failed!
Wilma Paxton Doherty is a Senior Consultant at ON THE MARK, a global organization design consulting firm and leader in collaborative business transformation with offices in the US and UK. She has over 15 years experience in consultancy, previously having held Board level roles in the public sector in the UK and a senior advisory role in a national public health organization.