The 5 Whys — drilling from symptom to systemic cause
The 5 Whys is the simplest root-cause tool and the easiest to do badly. Done right it walks you from a surface symptom down to a systemic gap; done lazily it stops at human error and calls it a day.
The 5 Whys is an iterative technique: you state the problem, ask why it happened, then ask why of each answer, repeating about five times until you reach a systemic root cause rather than a surface symptom. "Five" is a guideline, not a rule — you stop when the answer is something you can fix at the system level. Its weakness is that it follows a single chain, so it pairs well with a fishbone (which gives breadth) and with verifying each step against evidence.
How the chain works
You start with a clear problem statement and ask why did this happen? The answer becomes the new problem, and you ask why again. Each step should move from effect toward cause: short shot → cavity not filling → injection pressure low → check valve sticking → no preventive-maintenance schedule. Notice the destination: the first answer is a process symptom, the last is a systemic gap you can actually close.
How to do it well
1 · Five is a guideline
Some causes surface in three whys, some need seven. Stop when you reach something systemic — a missing standard, an absent control, a process gap — not when you hit an arbitrary count.
2 · End at a system, not a person
If a why lands on "the operator made a mistake," keep going: why was the mistake possible? A robust root cause is a system that allowed the error, because that's what you can error-proof. Stopping at blame fixes nothing.
3 · Verify each link
Each why should be supported by evidence, not assumption. If you can't confirm the link, you have a guess — and a wrong early link sends the whole chain off course.
4 · Branch when needed
If a why has two plausible answers, the single chain is too narrow — switch to a fishbone to hold the branches, then 5-Why the strongest one.
Common mistakes
- Stopping at human error. "Operator forgot" is a symptom of a missing safeguard, not a root cause.
- Unverified links. A chain built on assumptions reaches a confident wrong answer.
- Forcing a single line when the problem has several contributing causes — that's a fishbone, not a 5 Whys.
- Treating it as proof. The chain points you at a likely cause; confirm it before you act.
Where this gets slow by hand
Asking the questions is instant. Answering them with evidence is the work: each why needs a look at process data, maintenance records, or material history to confirm the link. Pulling that evidence at every step, across recurring problems, is where a quick exercise turns into a day of data archaeology.
A 5 Whys where each link is backed by data
Niobia supports each step of the chain with evidence — pulling the process traces, maintenance history, and material records that confirm or break the link between one why and the next, so you don't descend a chain built on assumptions. It fits the root-cause step (D4) of an 8D, and because every closed investigation becomes searchable memory, a recurring problem surfaces the chain that solved it last time rather than starting from the symptom again.
Frequently asked
What is the 5 Whys technique?
An iterative root-cause method: state the problem and ask why it happened, then ask why of each successive answer, about five times, until you reach a systemic root cause rather than a surface symptom. It was developed at Toyota as part of its production system.
Why is it called 5 Whys if you don't always ask five times?
Five is a rule of thumb for how many iterations it typically takes to get past symptoms to a systemic cause. You stop when the answer is something you can fix at the system level — sometimes that's three whys, sometimes seven.
What's the difference between 5 Whys and a fishbone diagram?
5 Whys follows a single chain of cause and effect to depth; a fishbone explores many cause categories in breadth. They complement each other — use a fishbone to surface candidate causes, then 5 Whys to drill into the most likely one.