Toyota Kata: Avoid Pareto Paralysis

A great key point comes out on page 124 of Toyota Kata by Mike Rother.

…a Toyota person once told me to focus on the biggest problem. However, when I tried to do this I noticed a negative effect: We got lost in hunting for and discussing what was the biggest problem. When we tried gathering data and making Pareto charts, it took a lot of time and the biggest problem in the Pareto chart was usually “other” which put us back into debating options. By the time we decided what the biggest problem was, the situation at the process had changed. This effect is called Pareto paralysis, and I encourage you to avoid it. Pareto paralysis delays your progress as people try to determine the “right” first step to take.

[…] it matters more that you take the first step than what the first step is.

I have seen this many times with my own eyes. It is a common problem when “problem solving” is taught as “application of the seven tools” and “correct” use of the tools becomes more important than understanding the problem or making actual progress.

Hirano, I believe, has been quoted as saying:

Waste often comes disguised as useful work.

That is what is happening here. The team thinks they are working to solve the problem when, in reality, they are still trying to decide what problem to work on. The gridlock may be driven by fear of working on the wrong problem. But in reality, there is no wrong problem, there is only the one that is in your way.

Here is something else I commonly see. When they discover “problem solving” many organizations want to immediately undertake large projects that involve “stretch goals” and breaking down complex issues. They look at operational metrics that are actually aggregated – total inventory, total lead time, productivity of the entire factory, total defect rates, total lead time through the system. While those metrics are nice to gage progress, they do not give you any actionable information.

Even worse is using averages. Averages aggregate even more, and “bad” tends to be countered by “good” in many average calculations. What is important is to look at each instance vs. an external, discrete, and objective standard or specification. Average defect rates tell you nothing at all about what to work on. But this unit had this defect, or this part dimension was this amount out of specification DOES.

Put another way, there are very few big problems. Big “problems” are the aggregated symptoms of many small problems.

This is good news and bad news.

The good news is that small problems can usually be understood, cleared and sometimes solved fairly quickly. The bad news is that it takes work to do this. The other bad news is that there isn’t any other way that actually works.

Doing a good job working on large complex issues requires engaging across the organization, time, and a high level of skill at understanding the situation, framing the problem, getting to causes, working on solutions, testing understanding, etc. Doing it well requires a mentor who is intimately familiar with the thinking behind problem solving.

This skill can not be gained in a “problem solving” or “A3” class, at least not one that simply walks the team members through the process once or twice. Skill comes from practice, and effective practice comes from starting simple, working in an environment where it is safe to experiment and to be wrong. This means working on shop floor issues that can be isolated from one another, the effect of a test or simulation can be seen immediately. The factory and production organization that Steven Spear describes in his research about Toyota is actually designed deliberately to facilitate this kind of work.

Another common mistake is setting up performance targets too early. If the processes are inherently unstable – if there is no sense of takt time, you are having problems with consistent quality or delivery, the first goal is simply to get to the point where you actually know how things are performing against an indicator of stability. Again, this is usually much simpler than setting a high level target and then trying to break down and stratify all of the issues out there. Set that stability target, study the process, and take on the disruptions as you see them. If there is a debate about which one to start with, then start with the simplest one so you can improve your capabilities.

In the end, it is about taking on problems as they come up, where they come up. That thinking is facilitated by the way you structure the flow, the work, and the organization itself. Get to the shop floor, stand in the chalk circle, and ask “What is stopping this team member from being successful right now.”

4 Replies to “Toyota Kata: Avoid Pareto Paralysis”

  1. Mark,

    I tend to see the problem with Pareto Paralysis (or and paralysis by analysis, for that matter) as a resul of not having clear goals. When the problem is not well defined, it is difficult to dive into. People view the problem differently, which makes them collect and review data differently, leading to the problems Rother mentions.

    That’s where the practice, as you mentioned, comes in, but it also requires a problem solving process. Without a process, there is nothing to practice.

    Hope all is well with you.

  2. I’m a fan of paretos, but I think they actually rarely work to help us find the biggest problem. The reason is that we can only put on the pareto the bins, or categories, of KNOWN problems. Data is pre-filtered for the most part based on the categories we determine. Very often, the biggest problems are unknown. They are not already categorized. It takes a different approach to find the biggest problems.

    Jamie Flinchbaugh
    Lean Learning Center

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.