Key facts
- Origin: Sakichi Toyoda, founder of the Toyoda companies; Taiichi Ohno made it a standard part of the Toyota Production System.
- Used for: finding the root cause of a problem that has one main chain of cause and effect.
- Who runs it: the people who do the work where the problem happened, led by a team leader, supervisor or engineer.
- Output: one root cause in the process, a countermeasure with an owner and a date, and a check that the fix worked.
- Time: usually 15 to 45 minutes, at the process or the team board.
What is the 5 whys technique used for?
The 5 whys is used to stop a problem recurring rather than just fixing its symptom. A machine stops, a batch fails inspection, an order ships late or a near miss is reported. The quick fix restarts the machine or reworks the batch; the 5 whys asks why it happened at all, and keeps asking until the team reaches a cause in the process that it can change.
It suits everyday, single-thread problems that a team can solve at its own level: breakdowns, defects, delays, rework and minor safety events. For complex problems with many possible causes, start wider with a fishbone or Ishikawa diagram and use the 5 whys on the likely causes.
How do you do a 5 whys analysis?
- Write a specific problem statement: what happened, where, when and how much, with a number in it.
- Go to the process with the people who do the work, rather than meeting in an office.
- Ask why the problem occurred and write the answer as a fact.
- Confirm the answer with evidence (an observation, a measurement or a record) before asking the next why.
- Repeat for each answer, keeping one clear chain of cause and effect.
- Stop at a systemic cause in a process, standard or system that the team can act on.
- Check the logic backwards using "therefore" from the root cause up to the problem.
- Put a countermeasure in place and verify it works over the following days or weeks.
To run one on paper or at the board, use the 5 whys template: a one-page worksheet with a problem statement, five why rows with an evidence column, and the root cause, countermeasure, owner and check.
What is an example of the 5 whys?
Each why takes the team one step from the symptom towards a cause it can fix.
Problem: Line 3 stopped for 40 minutes on the night shift because the main motor tripped.
- Why did the motor trip? It overheated under load.
- Why did it overheat? The bearing was running dry.
- Why was the bearing dry? The automatic lubricator was empty.
- Why was the lubricator empty? Nobody refilled it at the last scheduled check.
- Why wasn't it refilled? The lubricator is not on the autonomous maintenance checklist, so refilling depends on one technician remembering.
Countermeasure: add the lubricator level check to the autonomous maintenance checklist with a visual indicator, then confirm over the following weeks that the checks happen and the trips stop.
Read backwards: the lubricator is not on the checklist, therefore it was not refilled, therefore it ran empty, therefore the bearing ran dry, therefore the motor overheated and tripped. The logic holds.
How do you know you have found the root cause?
Use three tests. First, the chain reads backwards with "therefore" at every step; if one step does not follow, the chain is broken and needs another look. Second, the cause sits in a process, a standard or a system, not in a person: "the operator forgot" is where the next why starts. Third, removing the cause would have prevented this occurrence and will prevent the next one. If the countermeasure is only a reminder or a retraining session, the cause is probably still there.
5 whys vs fishbone diagram: which should you use?
| 5 whys | Fishbone diagram | |
|---|---|---|
| Direction | Deep: follows one chain of cause and effect | Wide: lists many possible causes by category |
| Best for | Straightforward problems with one main cause | Problems where causes are many or disputed |
| Output | One root cause and a countermeasure | A set of likely causes to test |
| Used together | Drill down on each likely cause | Run first, then 5 whys on the likely causes |
For large, cross-functional or repeating problems, both tools sit inside a structured method such as an A3 report or a CAPA record.
What are the common mistakes with the 5 whys?
- Stopping at blame: "the operator forgot" is not a root cause.
- Following guesses rather than facts verified at the process.
- Forcing a single chain when the problem has several contributing causes. Run a separate chain for each branch, or use a fishbone diagram.
- Ending on a cause outside the team's control, such as "the supplier is unreliable", without asking why the process is exposed to it.
- Counting to five and stopping. Some problems need three whys, others seven.
Sources
- ASQ: Five whys and five hows
- Lean Enterprise Institute: 5 Whys (Lean Lexicon)
- Ohno, T. Toyota Production System: Beyond Large-Scale Production (1988).