Was Deming against self-improvement?

I recently blogged about Adaptiv’s peer performance conversations, or PPCs. As a comment to that post, I received a great question from Bob Marshall (@flowchainsensei), which was: “Please tell me how this in any way makes sense in the context of Deming’s 95/5?”. This is my response.


I’m no expert on Deming but one of his most controversial statements was the so-called 95/5 rule. As I understand it, the rule suggests that 95 percent of the observed capability of a system (e.g. a business process) is determined by the design of the system, and only 5 percent is determined by the individual workers within that system. I interpret your question to mean, why focus on improving “the workers” when they only determine a very small percentage of the capability? I hope this interpretation is correct.

First of all, the purpose of the PPC is not primarily to improve the capability of a system, but to better ourselves as human beings (or not get worse). We can all improve, as individuals or as team members, don’t you think? I know I can. Sometimes I even lapse back into bad habits. When that happens, I’m very glad if I have colleagues who will remind me of how they want me to behave. The PPC is a good opportunity for that to happen.

You, or Deming’s ghost, might then argue that it’d be to no avail, since it would hardly affect the overall performance of the system. I believe this is true sometimes, but I can also think of several scenarios where this would be incorrect:

  1. It differs depending on type of process. I do believe most of Deming’s experience came from employing statistical methods to defined processes, like industrial settings. And it is a controversial statement. It’s not unreasonable to imagine that the proportions 95/5 could be different in other types of work, e.g. creative work, where the processes are empirical and mostly consist of humans jamming together. This is also the belief in the agile manifesto: Humans and interactions over processes and tools.
  2. The context sets the system capability, but we are far away from the capability limit due to other bottlenecks, e.g. poor teamwork or unbalanced distribution of experience. Yes, this is partly a system factor but partly also the responsibility of the team. By understanding what my peers want from me I can become a better team member and improve the team. I have experienced many times how a group of developers transforms into a team and then further into being a decent agile team, only then to hit system bottlenecks out of their control, e.g. pointy-haired managers, a ban from meeting users or outsourcing of operations.

Another thought I have is that we could use PPC on the management team. We have to remember that even if we are guided by our surroundings to a greater extent than we would like to admit, we are also the creators and maintainers of those surroundings. What if the other managers say this to their peer: “We have observed that you spend very little time observing how the work works. You are in charge of helping your people improve. How will you do this if you do not have intimate knowledge of their everyday struggles? We believe you should focus more on coaching change.” This is powerful feedback, which could lead to big ramifications.

Even if Deming was correct, he did not say that the performance of people did not matter. Even in organisations more or less run by the workers, e.g. Johnsonville Foods, they still demanded good performance from each other (at least not indifference). Perhaps a PPC every month is just about the level of attention we want to put on the individuals? The rest of the time we focus on better teamwork, better processes, improving capability. Call it an agile performance review, if you like.

To conclude: I don’t interpret Dr. Deming’s 95/5 as being against self-improvement. There is nothing wrong with trying to better one self. Rather I believe the rule should be interpreted as a wake-up call for management to stop looking only at the workers and start looking at the context; the major processes, structures, and systems within which work takes place. That’s their responsibility and I couldn’t agree more. But my performance (behaviour, attitude, focus) in the workplace is my responsibility and I need all the feedback I can get on how to improve it.

4 thoughts on “Was Deming against self-improvement?

  1. Great post.
    I would just argue about the infamous Deming’s quote

    Deming’s quote is controversial only because it is always reported incorrectly, and quoted as a “rule”, where it was in effect just little more than a comment, in which he just cited 2 arbitrary numbers, with no measurement nor statistic involved.

    In effect, there’s no 95/5 rule at all. Really.

    About the myth and the hyperbolic and literal interpretation of the 94% quote, could I suggest the post The Deification of Deming?

    It is worth a read

    http://www.leanway.no/the-deification-of-deming

Comments are closed.