Myth #3

Asking 5 Whys is an easy technique; just keep asking why, and you will always get to the root cause


February 2019


Johan Harder

As a concept, the 5 Whys is indeed easy to explain, and so naturally people are quick to start using it. But that is where people often go wrong. Just because it is easy to explain does not mean that it is easy to use … Correctly!

The architect of the Toyota Production System, Taiichi Ohno, described the 5 Whys as “the basis of Toyota’s scientific approach … By repeating why five times, the nature of the problem as well as its solution becomes clear”.

Indeed, 5 Whys is a seemingly straightforward technique. But I would argue that it is a myth that 5 Whys is an easy technique to use. Just because you keep asking “why”, does not mean that you will always get to the root cause of the problem. Or at least the “right” root cause. That has certainly been my own, rather painful, experience.

In order to know when and how to apply 5 Whys, it is critical to understand that the tool has certain limitations. Some are obvious, and some less so.

Let us start with the obvious limitations:
  1. The user’s current knowledge. If the root cause is beyond the user’s current knowledge, then the technique will not be of much value. Said differently, you need to ask the right people if you want to get the right answer. This is fairly obvious, but often the people directly involved in the work where the problem has occurred are not part of the analysis.
  2. Facilitator experience. If the facilitator is not experienced in the technique, often the “right” whys will not be asked. For example, if the facilitator does not clearly distinguish between symptoms and causes of a problem, questioning may start to focus on why a symptom of a problem has occurred. Then the analysis will just go in circles.
  3. Possibly multiple root causes. Users tend to focus on finding a single root cause for a problem, whereas, in reality, problems often have multiple root causes. If you only solve one of the root causes, guess what, the problem will continue to reoccur.
  4. Different perspectives. Lastly, users with different perspectives may identify completely different root causes to the same problem.

Those were some of the more obvious limitations to 5 Whys. Typically, users get wise to them after a couple of attempts at the technique. I certainly did.

There are, however, some less obvious limitations to which users often continue to fall victim.
  1. Confirmation bias. Like all other qualitative methods of analysis, 5 Whys is vulnerable to bias. When conducting an analysis, people have a natural tendency to search for, interpret, favour and recall information in a way to confirm their pre-existing beliefs or hypotheses. This is called confirmation bias. And guess what? These pre-existing beliefs are seldom correct. This risk is especially high when the technique is used in a change project where different agendas or perspectives invariably exist. The best way to mitigate against this risk is to back up the findings with data whenever possible.
  2. Observation vs deduction. Another limitation, or pitfall rather, is to make the 5 Why analysis a deductive exercise conducted in a classroom far away from the action. 5 Whys was developed with Toyota’s “Go and See” philosophy in mind where the observations are intended to be the facts of the analysis. When used in a classroom setting, the technique is used in a deductive manner rather than being based on observation. Furthermore, when it becomes a classroom exercise, the user is separated from the environment of the actual process (the action), and the exercise naturally loses its rigour.
  3. Fact vs fiction. The last of the less obvious limitations is that 5 Whys can and should only be used on a problem that has actually happened, and preferably only that one factual occurrence. I have often experienced people suggest the 5 Whys as a technique to hypothesise about why a problem might have occurred, or, even worse, why it could occur in the future. I was once part of a workshop where the tool was applied to understand why a person could be late for an appointment. That is simply not the intended use of the tool. It is not for fiction.

That was some of the main limitations of the 5 Whys. The obvious, and not so obvious, limitations to which people often fall victim. Does this confirm that it is a myth that the 5 Whys is a straightforward root cause analysis technique? Maybe, maybe not. But what I can confirm is that I certainly wish that I had been aware of these limitations before I started using the technique.