The purpose of measuring any process is to uncover variation. After all, if something doesn’t vary, you don’t need to measure it. However, what we want to understand is the causes of that variation, so we can figure out what changes will achieve our goals. Often, process improvement experts will talk about two kinds of causes:
- Common Causes, which account for the normal behavior of the process
- Special Causes, which account for changes in the behavior of the process
Trevor, a Lean Six Sigma black belt at a plant within Danaher Corporation, asked me a valuable question about this distinction:
“You say in “Sales and Marketing the Six Sigma Way” that managers need to focus on special cause variation but they should not really be concerned with common cause variation.
“I agree that we need to understand significant instances of special cause variation, but I feel strongly that much of the improvement opportunity lies within focusing on the fat that often lies within the common cause variation…
“Most folks don’t know what inputs (special or otherwise) cause their variations… yet only once they do can they begin driving improvement. Agree?”
Yes, I agree. Trevor has made a helpful distinction.
The comment in the book about “special causes” was made to point out that most executives hurt themselves when they do not understand how variation works, such as the difference between common and special causes.
As a result, if sales figures are slightly lower this quarter than last year in this quarter, they may have a knee-jerk reaction to change something, on the grounds that the results are lower than last year.
Of course, all this does is make things worse, because, to Trevor’s point, if they have not analyzed their process, they do not know the cause of what they are experiencing, or whether it results from common or special causes. It is no wonder they have so little chance of improving anything.
The whole idea of Six Sigma is to break things down so that analysis of measured data can help identify what is really happening. Combining measured data with knowledge of the context enables people understand which inputs are causes of which types of variation in the outputs.
When a process is “in control” (i.e., its variability is understood and predictable) it is easier to detect a “special cause,” i.e., when something changes the behavior of the process or system. In a noisy process, as is usually the case in sales and marketing, it can be difficult to identify a special cause.
If you need to improve the performance of the process generally, and there is no identifiable “special cause” you can easily fix, then you are in a position of, as Trevor put it, “focusing on the fat that often lies within the common causes of variation.”
This is exactly where Lean and Six Sigma techniques are most valuable. These principles that enable you to think through the process and the problems in a systematic way, so you can ask questions and consider alternatives that traditional thinking overlook.
If you’d like a guideline for applying this kind of thinking you your process, you’ll enjoy “Sales and Marketing the Six Sigma Way.”
October 23, 2006