Why? Why? Why? Why? Why?
©2020, George J. Irwin. All rights reserved.

We now turn to the tool that makes every Lean Six Sigma Practictioner sound like an impatient three-year old... one of the several approaches to Root Cause Analysis commonly known as the "Five Whys."

The idea is deceptively simple: keep asking "Why?" until at least one, well, root cause, is determined. Here's an example:

The power is out and my flashlight doesn't work. Why?

The flashlight battery is dead. Why?

I didn't replace it when it died. Why?

I didn't know it was dead. Why?

I don't have a plan to test flashlights when I have power instead of when I need flashlights.

The quanitity five is somewhat arbitrary here. It could take more "whys" than that. If you've stopped at just one, two or even three "whys" than in my experience you've either hit an "Existing Reality"—something for another post—or, sorry, you're not quite thinking things through enough.

What's important here is that the exercise results in something actionable. In this case, the countermeasure is to have a plan to test flashlights when you don't need them. I could have come up with a Why about not having spare batteries on hand, but if I'm already in the dark without a flashlight, I'm not going to be able to see where they are. I could have asked another Why about not having a plan, but that would probably be answered with a generality like "I don't have time" or "I didn't think of it" or, in my case, "I'm lazy." None of these are as easy to address as simply having a plan to test flashlights. (Actually executing that plan, of course, is another matter entirely.)

Five Whys in the Real World are typically not that simple. There could be multiple paths and multiple causals; keeping track of them requires anywhere from a single sheet of paper to a large wall and legible writing.

A robust analysis can point to multiple process weaknesses. When I was on a project to understand why expert call center representatives couldn't regularly "clear the queue" of incoming inquiries, we had something like two dozen root causals. We asked "Why?" a lot that day. Fortunately, we were also able to group some causals together, using something called an Affinity Diagram, and since we knew we couldn't attack all of the root causals at once, we pulled out some prioritization techniques to determine which ones to address first.

Things can also let a little sensitive with this tool. In the call center example, one of the answers to "Why" was "We don't have enough representatives" and the answer to the succeeding "Why" was "Management doesn't want to hire any more people." Oops! Even if that turns out to be an Existing Reality, it's important to at least recognize what might turn out to be unpopular root causals.

More from the Notebook...