5 Whys: How to find the root cause?
5 Whys is a powerful technique that can be used to identify the root cause of problems faced by your business. It was initially introduced by Sakichi Toyoda, one of the founders of the Toyota Group of companies. Toyota used this to find problems in the functioning of its business making them the largest automobile manufacturer in the world.
The process involves you keep asking why 5 times recursively. Ask why first, then ask why on the answer of the first question, then ask why on the answer of the second question, and so on. The final answer should be the root cause of the problem.
After this analysis, we have to decide on the solutions. Each of the problems must be solved in such a way that they should not occur again. Let’s take an example. The problem that occurred in the following case was that a website was down for some time.
1. Why was the website down?
The CPU utilization on all our front-end servers went to 100%
Solution: Bring the site back up
2. Why did the CPU usage spike?
A new bit of code contained an infinite loop!
Solution: Remove the bad code
3. Why did that code get written?
Person X made a mistake
Solution: Help person X understand why his code doesn’t work as written
4. Why did his mistake get checked in?
He didn’t write a unit test for the feature
Solution: Train person X in the principles of Test Driven Development
5. Why didn’t he write a unit test?
He’s a new employee, and he was not properly trained in Test Driven Development
Solution: Change the new engineer orientation to include TDD
The number 5 has nothing special in it. The general rule of thumb is to keep questioning until you stop getting actionable answers. Answers that are impossible or unpredictable are usually signs to change the answer or stop the asking process entirely.
So now I can finally introduce the dark side of the 5 Whys — The 5 Blames.
It is a state where we just blame everyone else for the problem instead of finding the cause. It usually happens when we overreact to problematic situations . So this process of finding and fixing problems can instantly turn into a means for venting frustration on other people. A way to prevent such situations is to do a 5 why analysis in a group instead of doing it individually. The group should contain all the members that were involved in the problem. Even people originally found the problem and those who had attempted to fix it in the past. It’s also important for everybody from the lowest rank to the highest rank to be present. The reason to do so is that generally, whoever is not at the meeting ends up being the target of the blame. Let me end it with a quote from Eric Ries’s book “The Lean Startup”:
“Although it’s human nature to assume that when we see a mistake, it’s due to defects in someone else’s department, knowledge, or character, the goal of the Five Whys is to help us see the objective truth that chronic problems are caused by bad process, not bad people, and remedy them accordingly.”
The following are the few more drawbacks of 5 whys:
- This process cannot find causes that we do not already know. In the above example if we don’t know what a unit test is then there is no point in continuing further. Hence the lack of knowledge makes a 5 why analysis useless.
- People tend to find one root cause for every problem but reality could not be more incorrect. There can be two or more causes for each problem and we have to solve all of them. Each of the individual causes will itself have two or more causes. Finally, the whole thing will look similar to this super branched structure that in computer science, we refer to as a tree.
- Making a mere linear list of 5 whys is often not enough. Sometimes we have to make this complicated 5 why tree with the number of causes growing exponentially.
- The 5 why analysis is not repeatable. If 10 people were given the same problem and then performed the whole process, we would get 10 completely different answers by the end. It is best to not even let 10 people run this process separately. There is a reason why an analysis by a group leads to more insight than by a single individual.
Credits : Elbert Ronnie — Team E-Cell ABV-IIITM Gwalior