Incident management for high-velocity teams
In defense of 5 whys
A few years ago, 5 whys was the next big thing. Now, in many circles, it’s fallen out of favor. Has the practice really gone belly-up, or are we throwing the baby out with the bathwater?
In the 1930s, Toyota founder Sakichi Toyoda came up with a simple way to get to the root of an issue: Identify the problem and ask why it happened. Then ask why that happened. Then ask why that next thing happened. And so on until you get to the true root cause of your problem. He called it the five whys, because usually he’d land at the problem’s root cause by the fifth question.
Come the 1970s, his simple method took off, gaining traction in not only manufacturing, but other industries—including tech—that needed a simple way to get to the heart of incidents and issues.
These days, Toyota still uses the 5 whys to diagnose causes. But for many forward-thinking companies, the romance is fading, with some suggesting the methodology has run its course.
But has it? Is the 5 whys really irrelevant now? Or are we throwing out the baby with the bathwater?
Here are the five most common objections to the methodology—and why we don’t think these 5 whys hold up to scrutiny.
Objection: 5 whys contributes to blame culture
At best, this objection is a misunderstanding of how the 5 whys and blameless culture work. Identifying a problem isn’t the same as shaming those who had a hand in it. And taking accountability without fear of personal consequences is a core value of good blameless postmortems.
One of the foundational rules for 5 whys is never to identify a person or people as the root cause. Human error, team B’s screwup, and lack of employee X attention aren’t acceptable answers to the question. What caused the human error? What led to team B’s screwup? What process, alterable behavior, or cultural factors are at play here? Those who stop at a place of blame haven’t gone far enough with their whys.
Objection: 5 is too many/too few
The 5 whys isn’t meant to be literally 5 whys. Five is shorthand for a process that pushes you to dig deeper until you hit the broken process or processes behind the incident.
Sometimes you’ll get to your answer with three whys. Sometimes it’ll take more than five. The point here is to keep going and take a deeper analysis of your cause-and-effect chain.
Objection: there’s rarely a single root cause
There’s no hard and fast rule that the answer to your final why has to be a single cause. There’s no 5-whys police coming for you if your cause-and-effect chain turns into a tree of multiple factors.
The goal here is to use a simple question—why?—to get to the heart of what went wrong. More complex problems may require additional approaches, but that doesn’t negate the power of a simple string of whys.
Objection: it’s not holistic enough
Isn’t it, though? One of the rules of the framework is that the heart of the issue is never a tech glitch or “it’s Jack’s fault!” The heart of the issue is a broken process.
Why is it Jack’s fault? We didn’t adequately prepare Jack for success. Why did that tech glitch happen? We have a process that allows us to launch changes without proper tests in place to catch the issue.
5 whys isn’t just about identifying a string of technology causes-and-effects. It can just as easily incorporate process, culture, and big-picture business setup as long as your team isn’t ignoring those factors.
Objection: we can’t know what we don’t know
Investigators can only identify causes when they understand the intricacies of those causes. A common objection to the 5 whys is that this isn’t good enough. Our investigators can’t know what they don’t know. A single person can’t possibly correctly find the cause-and-effect chain, no matter how many times they ask why.
To this, we say the answer is that 5 whys is a team effort. Everyone involved in the incident should be involved in its postmortem. People with hands-on experience need to be in the room. Management—bringing another perspective to the table—needs to be involved. This isn’t a process that always works in a vacuum, but that doesn’t mean it doesn’t work when you give everyone a seat at the table.
5 whys—yay or nay?
So, is the 5 whys the perfect solution to every problem? Of course not. There’s more than one way to get to the heart of an incident. Some internal cultures might be better suited to another approach. And there are certainly ways to misuse or misunderstand the 5 whys and end up with an unsatisfactory result.
But that doesn’t mean the practice is irrelevant. across the board. In fact, we’d argue that for many it’s still a useful tool for digging past symptoms and seeing the big picture. As far as we can tell, the big objections to the methodology all miss the point or misinterpret the practice.
So, is the 5 whys methodology dead in the water? We think the answer is “only if you’re doing it wrong.”
Learn incident communication with Statuspage
In this tutorial, we’ll show you how to use incident templates to communicate effectively during outages. Adaptable to many types of service interruption.
Read this tutorialThe importance of an incident postmortem process
An incident postmortem, also known as a post-incident review, is the best way to work through what happened during an incident and capture lessons learned.
Read this article