The 5 Whys Method: A Guide with Practical Quality Management Examples

Updated 30 September 2026. This article has been revised and now rests on primary sources: Taiichi Ohno’s own description of the method, Alan J. Card’s critical analysis in BMJ Quality & Safety, ISO 9001, IATF 16949 and Ford’s customer-specific requirements. The former Wikipedia source has been dropped.

Asking “why?” five times — hardly any quality method sounds more trivial, and hardly any is misapplied so reliably. The principle is rarely the problem; the problem is that the chain stops at the first comfortable answer, usually a person. This article shows what a chain that holds up looks like, how to recognise one that stopped too early, and where the method reaches its limits.

Origins: Ohno’s machine

The method is tied to the Toyota Production System that Taiichi Ohno built. In his book on the Toyota Production System, Ohno describes repeatedly asking why as the basis of Toyota’s scientific approach and illustrates it with a machine that has stopped [1]. In paraphrase:

1. Why did the machine stop? An overload blew the fuse.
2. Why was there an overload? The bearing was not sufficiently lubricated.
3. Why was it not sufficiently lubricated? The lubrication pump was not pumping enough.
4. Why was it not pumping enough? The pump shaft was worn.
5. Why was it worn? There was no strainer, and metal chips got into the pump.

The fix is a strainer — not a new fuse, which would have “solved” the problem after the first answer [1]. That is the core of the method: not the number five, but refusing to stop at the first answer that explains the symptom.

The chain: every answer is the next question

The most widespread mistake is treating 5 Whys as a form with five lines. It is a chain: each answer becomes the next question, until the analysis reaches a cause the organization can change itself. Sometimes that takes three steps, sometimes seven.

Two checks make a chain hold up:

• Read it backwards. Joined bottom to top with “therefore”, the chain must still make sense: no strainer, therefore chips in the pump, therefore a worn shaft, and so on. Where the sentence breaks, a step is missing.
• Evidence every answer. “The bearing was not lubricated” is a guess until someone has looked at the bearing. A chain of plausible but unchecked answers looks finished and leads to the wrong action.

Assessment: the second check is the one most often skipped, because 5 Whys tends to happen at a desk rather than where the problem occurred.

Three examples

Manufacturing: a milled bore out of tolerance

1. Why was the bore out of tolerance? The tool was worn.
2. Why was a worn tool in use? The tool change was not triggered.
3. Why not? It is tied to a piece count, not to actual tool life.
4. Why a piece count? The tool data sheet was never updated with real wear data after series launch.
5. Why not? There is no process that reviews tool data periodically after launch.

Cause: a missing feedback loop between series production and tool data maintenance — not carelessness at the machine.

Administration: an invoice two weeks late

1. Why was the invoice late? Completion of the work was not recorded in the system.
2. Why not? Project managers do not report completion; accounting looks for it.
3. Why does accounting look for it? There is no defined handover between project completion and invoicing.
4. Why not? The process was designed for a few individual projects where accounting knew informally anyway.
5. Why was it not adapted? Nobody owns the invoicing process.

Cause: a process without an owner that did not grow with the organization.

Supply chain: wrong part variant delivered

1. Why was the wrong variant delivered? The supplier built to an outdated drawing.
2. Why did it have the outdated drawing? New revisions are not provided to it automatically.
3. Why not? Drawing changes go out as email attachments.
4. Why by email? There is no interface between PLM and the supplier portal.
5. Why not? It was classed as “can be added later” when the supplier was onboarded and never prioritised.

Cause: a missing technical interface, not a careless supplier. The example also shows a practical rule: “because the supplier changed its process” is a fact, but not yet a cause your own organization can act on.

The most common mistake: the chain ends at a person

A reliable warning sign is a last answer that points at a person — “operator was not paying attention”, “supplier was careless”. Human error is almost never the cause where something can be changed for good; it is the point where the analysis became uncomfortable. The next question then is: why could the process let this error through without catching it?

What ISO 9001 and IATF 16949 require

ISO 9001 clause 10.2.1 b) requires the nonconformity to be reviewed, its causes determined, and a check made whether similar nonconformities exist or could occur [3]. The standard prescribes no method; 5 Whys is one of several acceptable ones. The question about similar cases is the one most often missing from 5 Whys chains — it belongs at the end of every analysis. ISO 9001:2026 keeps this structure [4].

IATF 16949 goes further: clause 10.2.3 requires a problem-solving process that prevents recurrence, with root cause analysis including the methodology used, the analysis and the results, and systemic corrective actions that also consider similar processes and products [5]. The method has to be documented, not just the result.

Customer requirements can prescribe a method. Ford’s customer-specific requirements call for nonconforming product and process output to be analysed using 8D and “3x5 Why” [6].

Where the method reaches its limits

The most thorough critique comes from patient safety. In BMJ Quality & Safety, Alan J. Card argues that 5 Whys reduces complex events to a single linear causal chain, that different people analysing the same event reach different root causes, and that the method is unsuitable for analysing serious incidents [2].

Assessment: for quality management in manufacturing the critique is fair but not fatal. 5 Whys suits contained deviations with a recognisable, linear cause. For customer complaints, safety-relevant defects, several parallel causes, or when it is unclear why the defect was not detected, a team process such as the 8D report is the better fit — its D4 step consists of two separate why-chains, one for occurrence and one for non-detection.

From cause to action

A root cause analysis is not an end in itself. ISO 9001 distinguishes the correction, which deals with the specific nonconformity, from the corrective action, which removes its cause [3]. Without a chain carried through to the end that distinction is hard to draw — and the effectiveness of a corrective action can only be reviewed if it is clear which cause it was meant to remove. The flow from finding to effectiveness review is covered in the articles on the CAPA process and on audit findings and CAPA.

How qportal handles the chain

In qportal, a 5 Whys analysis is a guided chain rather than a free-text field. Each step is its own line with sequence, question and cause, optionally assigned to one of six cause categories (man, method, machine, material, environment, measurement). An analysis can be started straight from an audit finding. The final cause leads to actions with an owner, a due date and a lifecycle of their own. The analysis is only closed after submission for the effectiveness review and release by a root cause analysis manager, who can also send it back with a note. Details are on the 5 Whys page; the steps are in the 5 Whys documentation.

Conclusion

5 Whys is the most accessible root cause method and precisely for that reason the one most often applied superficially. Carry the chain to a cause the organization can change, evidence each step rather than assuming it, and ask about similar cases at the end — and you get a corrective action that holds. Know the method’s limits, and you also know when a team and two chains are needed.

Sources

1. Ohno, Taiichi: Toyota Production System: Beyond Large-Scale Production. Productivity Press, Cambridge, MA, 1988; Japanese original 1978. Consulted for the description of asking why five times and the machine example, paraphrased.
2. Card, Alan J.: “The problem with ‘5 whys’.” BMJ Quality & Safety 26 (2017), no. 8, pp. 671–677. https://doi.org/10.1136/bmjqs-2016-005849 (metadata checked 2026-09-30)
3. ISO 9001:2015, Quality management systems — Requirements, consulted in the bilingual edition DIN EN ISO 9001:2015-11, Beuth Verlag, Berlin. Clause 10.2.1.
4. ISO/DIS 9001:2025(en): Quality management systems — Requirements. ISO/TC 176/SC 2, committee ballot 2025-08-27 to 2025-11-19. Not publicly available; checked against a copy on 2026-09-30 (clause 10.2.1) — not the published IS edition.
5. IATF: IATF 16949:2016 — Sanctioned Interpretations (SIs). Status November 2025, including SI 20 on clause 10.2.3. https://www.iatfglobaloversight.org/wp/wp-content/uploads/2025/11/IATF-16949-SIs-Nov-2025.pdf (accessed 2026-09-30)
6. Ford Motor Company: Customer Specific Requirements for IATF 16949:2016. Effective June 2026, section 10.2. https://www.iatfglobaloversight.org/wp/wp-content/uploads/2026/06/Ford-IATF-CSR_Jun-2026-SCCAF_Final.pdf (accessed 2026-09-30)

Consultancy, certification-body and content-marketing pages are not cited. Normative text is copyrighted; requirements are paraphrased with their clause reference. Where this text goes beyond the documented evidence, it is marked as an assessment.