8D Report: The Eight Disciplines Explained

Most published guidance walks D1 to D8 in order and stops there. That misses what actually makes the method different from every other investigation format: 8D runs two tracks at once, looks for two causes rather than one, and is the only common method that requires you to take something away at the end.

What is an 8D report and where did it come from?

The method was developed at Ford Motor Company. Executives in the Powertrain organization wanted a way for design engineering, manufacturing engineering and production teams to work on recurring chronic problems together, staff were assigned to build training materials in 1986, and the resulting manual, *Team Oriented Problem Solving*, was first published in 1987 and piloted at Ford World Headquarters in Dearborn, Michigan. Ford’s current variant is called Global 8D, usually written G8D.

There is an older ancestor, and the relationship is disputed rather than proven. The US Department of Defense issued MIL-STD-1520, Corrective Action and Disposition System for Nonconforming Material, which set out containment of nonconforming material followed by corrective action. The commonly cited account dates its development to 1974; the revision that survives in full text is MIL-STD-1520C of 27 June 1986, which superseded MIL-STD-1520B of 3 July 1980, and the standard was cancelled in February 1995. Whether Ford drew on it directly is not documented; the structural resemblance is obvious and the causal claim is not evidenced. What that standard still supplies, in daily use, is the disposition vocabulary described on the non-conformance report page.

The method is now written into industry standards rather than owned by one manufacturer. SAE AS13000, Problem Solving Requirements for Suppliers, published by the Aerospace Engine Supplier Quality group, uses 8D as its basis and requires suppliers to hold a documented problem-solving process meeting the standard. ASQ publishes its own 8D reference. The label on the form is less important than the fact that a customer can ask for it by name and expect the same eight things.

The two tracks: protect while you investigate

8D is not a linear sequence, and reading it as one is the mistake that costs accounts.

It runs two tracks in parallel. A protection track, D0 and D3, buys time: it quantifies the symptom, puts an emergency response in front of the customer, and contains the affected material. A causal track, D2 and D4 through D7, spends that time: it describes the problem with data, finds and verifies the cause, proves a fix and prevents recurrence.

Teams that treat the disciplines as a queue run only the causal track. They arrive at a correct, well-evidenced root cause in week four and present it to a customer who has been receiving defective parts throughout. A contained problem is an internal problem. An uncontained one becomes the customer’s problem, and the customer’s problem is the one that costs you the account.

The two tracks also answer different questions, which is why they can run at the same time without interfering. Containment answers “who is exposed right now.” Investigation answers “why did this happen and why did nothing catch it.” Neither answer substitutes for the other, and a report strong on one and silent on the other is half a report whichever half is missing.

Is D0 part of 8D or not?

It depends whose 8D you are running, and the answer is worth checking before you fill in a template.

Classic 8D has eight disciplines, D1 to D8. Ford’s Global 8D adds a ninth step numbered zero. The published Global 8D workbook lists “D0 Prepare for the Global 8D Process” as a real step whose job is to define and quantify the symptom, identify who is affected, “provide an Emergency Response Action to protect the customer,” and then decide whether the problem warrants the effort of a full 8D at all.

That last function is the one people miss. D0 is a gate, not a formality: it is the only point in the method where the correct answer is allowed to be “this does not need an 8D.”

So the count is eight disciplines and nine steps, and a form with a D0 box on it is a Global 8D form.

What does each discipline have to produce?

The disciplines are easy to list. What matters for audit, and for whether the report is worth anything later, is the evidence each one leaves behind.

Discipline names follow the Global 8D workbook. Where a form says “Congratulate the Team” at D8 rather than “Recognize Team and Individual Contributions,” it is a simplification of the original, and a consequential one.

Why is D3 the discipline that decides the commercial outcome?

Because containment is the only discipline whose value decays by the hour.

D4 through D7 can be done well in week three and still be worth doing. An interim containment action implemented in week three is worth almost nothing, because by then the affected material has already reached whoever it was going to reach. This is why customer requirements put a clock on containment and nothing else. Aerospace supplier requirements published by Barnes Aerospace, which require 8D in the AS13000 format, state that containment “must commence immediately with response of containment results within 24 hours and corrective action plan within 30 days.” Twenty-four hours against thirty days is the whole argument in two numbers.

Containment is also expensive by design. The Global 8D workbook is blunt that an interim action “will be expensive to implement as it will probably involve an additional process,” and lists 100 percent inspection, a substitute process, rework and field service action as typical choices. It also notes that none of them is fully effective, because “even 100% inspection is subject to variation.” That is the trade the discipline exists to make: known cost now, in exchange for time to find the cause.

Where do 5 Whys and Ishikawa fit in?

Inside D4, in most renderings. The commonly cited description of D4 has the team identify all applicable causes, verify each one, and use tools such as five whys or Ishikawa diagrams to map causes against the observed effect.

But the sources genuinely disagree about placement, and the disagreement is informative. Ford’s Global 8D workbook puts the Is/Is Not analysis at D2 for problem description, uses comparative testing of possible causes against that description at D4, and then applies 5 Whys at D7 to reach what it calls “the root cause of the root cause,” the policy or practice that allowed the technical cause to exist. The Wikipedia rendering also attaches 5W2H, the who, what, where, when, why, how and how many framing, to D2, which explains why “5W2H method” is searched alongside 8D at all.

The resolution is that 8D specifies outputs, not tools. D4 requires a verified cause, and verification means the cause has been tested against the data rather than agreed in a room. A fishbone diagram with twenty-five branches and no test is not D4 output. One tested hypothesis is.

D4 also carries the requirement that distinguishes 8D from almost every other investigation format: the escape point. The team must isolate and verify “the place in the process where the effects of the root cause should have been detected and contained but was not.” That is a second cause, with its own corrective action at D5, and a report that names only the occurrence cause has completed half of D4. It is also the point at which the detection field on the originating record earns its place, because where a defect was found is the starting evidence for where it should have been found.

What does “verify” mean at D5 and “validate” mean at D6?

The two words are not synonyms in this method, and getting them the wrong way round is the most common technical error in 8D reports.

Verification happens before implementation. At D5 the team confirms, on a small scale or under controlled conditions, that the proposed corrective action would actually resolve the cause. Validation happens after. The Global 8D workbook defines it precisely: validation is “proof that is developed after implementation has taken place, over a period of time,” it “must follow successful verification,” and it must “provide a data driven comparison before and after implementation” and check that the change has not introduced a new problem.

Verification asks whether the fix would work. Validation proves it did, using the same measure that described the problem in the first place.

D6 is also where the containment comes out. The workbook instructs the team to “plan to remove the Interim Containment Action (ICA) as part of implementation of the PCA,” and this is the step templates most often drop. An interim action left in place after the permanent fix lands becomes an undocumented, uncosted permanent process, and quality functions accumulate these for years without noticing. 8D is the only common problem-solving method that requires you to take something away, which is exactly why that requirement gets edited out.

How is an 8D different from a CAPA or an NCR?

They sit at different points and answer different questions.

A non-conformance report is a custody record for affected material: what was found, what was done with it, who authorized that. A CAPA is a management-system obligation to eliminate the cause of a nonconformity and to check the action worked. An 8D is a method and a deliverable, usually produced because a customer asked for one, that carries a team through the investigation and hands over a single auditable document.

Most non-conformance reports close on disposition and never become an 8D. When one should escalate, and on what triggers, is set out on the non-conformance report page.

Mapping 8D onto a regulated CAPA system

If you work under US medical device rules, start with the date. The FDA’s Quality Management System Regulation took effect on 2 February 2026, replacing the old 21 CFR 820.100 CAPA section by incorporating ISO 13485:2016 by reference. Guidance that still cites 820.100 as current, including a good deal of published 8D material, is out of date.

The mapping trap itself is older and survives the change. Consultants working in FDA-regulated industry warn that “8D places corrective action verification ahead of implementation rather than after it”, while regulated CAPA requires verification that actions were effective once implemented. Read against 8D’s own vocabulary the gap closes, because the after-the-fact proof lives at D6 validation rather than D5 verification. But a quality system that maps 8D onto CAPA field by field, with D5 in the box labeled verification of effectiveness, produces a compliance finding from a method that did the work correctly.

In automotive, the pull comes from IATF 16949 clause 10.2.3, which requires a documented problem-solving process covering containment, interim actions, root cause analysis and verification of effectiveness. The standard does not name 8D. Customer-specific requirements from individual manufacturers frequently do.

Why is D8 the discipline everyone skips?

Because almost every template mislabels it. Rendered as “congratulate your team,” D8 reads like an optional courtesy, and under deadline pressure a courtesy is the first thing to go.

In the original method it is the closure discipline. The Global 8D workbook heads the section “completing unfinished business” and lists finalizing and archiving documentation, recognizing technical lessons learned, presenting to the champion, and “ensuring all other related documentation is revisited and up to date (e.g. FMEAs, Control Plans, etc.).”

That is not a thank-you. That is the step where the investigation’s findings get written back into the FMEA and the control plan, which is the only mechanism by which an 8D changes what the organization does next time. Skip D8 and the corrective action holds on one line, on one part number, until the person who implemented it leaves.

The recognition element is not decorative either. 8D is a team method that depends on people volunteering unwelcome information about processes they own, and the workbook’s reasoning is simply that recognizing the effort “tends to reinforce behaviour and self-esteem.” An organization that never closes the loop socially gets slower, thinner 8Ds over time, and the cause of that is invisible in every metric a quality function keeps.

Where 8D reports go stale, the failure is almost never in the middle. It is D0 skipped so the team never decided whether the problem deserved the method, D3 skipped so containment arrived after the customer did, and D8 skipped so nothing upstream changed. The disciplines in between are the ones everybody remembers to do, and internal quality audits are usually where the missing ends surface.

Frequently asked questions

Sources

• Wikipedia, Eight disciplines problem solving
• Global 8D Problem Solving Workbook, Results Consortium Limited
• ASQ, The Eight Disciplines of Problem Solving (8D)
• SAE AS13000, Problem Solving Requirements for Suppliers
• AESQ, 8D Problem Solving Method subject matter interest group
• Barnes Aerospace, Supplier Quality Requirements SQR-001, Revision K, 26 June 2023
• MIL-STD-1520C, full text, 27 June 1986
• MIL-STD-1520C Notice 2, cancellation, 27 February 1995, EverySpec
• The FDA Group, The problem with 8D for CAPA under FDA and ISO regulations
• FDA, Medical Devices: Quality Management System Regulation, final rule, Federal Register, 2 February 2024
• IATF 16949 clause 10.2.3, problem solving
• ISO 9001:2015, Quality management systems, Requirements

AS13000, IATF 16949 and the ISO standards are copyrighted and sold, so their requirements are described rather than reproduced and are cited by standard and clause number. The Global 8D workbook is a published training document reproducing Ford’s Global 8D structure; it is not a Ford publication, and discipline wording taken from it is quoted as such. The 1974 date for MIL-STD-1520 is the date given in secondary accounts of the method’s history; the earliest revision available in full text is the 1986 C revision.

Related terms

• Non-conformance report
• What is a non-conformance?
• CAPA explained
• FMEA
• Internal quality audits

Last reviewed: 16 September 2026

About Logincident. Logincident is a data and software company whose configurable platform captures structured evidence at the point of work and presents it in dashboards and reports, including quality records such as containment actions, root cause findings and corrective actions. We are not a law firm or a claims handler, and nothing on this page is legal advice.