How do chemical suppliers write an 8D report, with an example?
Summary
A chemical supplier's 8D report covers 8 disciplines; the hard ones are containment, root cause, corrective action and verification, which need lot-level evidence. That evidence means retained samples, stock counts by lot, Certificate of Analysis (CoA) data and batch records, and the report holds when the complaint can be traced to one or more delivered lots.
Who reads a chemical supplier's 8D report, and what do they decide?
A chemical supplier's 8D report is read first by the customer's supplier quality engineer, who decides whether the suspect material is under control, whether the stated root cause is proven, and whether the complaint can be closed. Behind that engineer sit the customer's production team, who need to know which lots they can keep using, and the customer's own auditors, who may later ask how the complaint was handled.
Automotive customers in particular read the 8D against their own certification duties. IATF 16949:2016 (1st edition) is the automotive quality management standard that replaced ISO/TS 16949:2009 (IATF Global Oversight, accessed 2026). Under its clause 10.2.3, organizations seeking certification are required to have defined approaches for containment, root cause analysis, implementation of corrective actions and verification of their effectiveness, as summarized by a certification body (BSI, accessed 2026). A supplier's 8D report becomes part of that customer's evidence, so it is judged by the same headings.
The 8D method itself is described in public guides. The VDA volume "8D - Problem Solving in 8 Disciplines", 1st edition, November 2018, covers the method, process and report, and states that it does not describe a general complaint process (VDA QMC, 2018). The AIAG CQI-20 Effective Problem Solving Guide, 2nd edition, is organized by problem-solving stage, with progress check questions for each (AIAG, accessed 2026). Neither guide is written for chemicals, which is why a chemical supplier has to supply the lot evidence the forms leave implicit.
What goes in each discipline of an 8D report from a chemical supplier?
Each discipline of an 8D report from a chemical supplier has to show one decision and the lot-level evidence behind it, from the team in D1 to closure in D8. Customer forms number and name the disciplines slightly differently; the table below follows the common D1 to D8 sequence and decides whether the customer will accept the chemical supplier's 8D report.
Discipline | What it must show | Chemical supplier evidence |
|---|---|---|
D1 Team | Who works the complaint, including the area where the fault arose | Names and roles from quality, production, laboratory and, where needed, the raw material supplier |
D2 Problem description | The customer's complaint restated and quantified | Product, lot numbers, quantities, dates, customer test values beside CoA values |
D3 Containment | Suspect material found and held at every location | Stock count by lot at the plant, in transit and at the customer; hold records |
D4 Root cause | Why it occurred, why it was not detected, which system gap allowed it | Retained sample retest, batch record, raw material lots, process data |
D5 Corrective actions | Actions chosen for each proven cause | Cause chain line each action addresses |
D6 Implementation and verification | Actions done, with success criteria set beforehand | Dated records; criteria fixed before the action ran |
D7 Prevent recurrence | Changes to the system: specifications, release tests, procedures | Revised documents with revision numbers |
D8 Closure | Customer informed, team closed | Final report date and sign-off |
The key row is D4: a chemical supplier's root cause is accepted only when the retained sample and batch record, not the narrative, show it.
D1, D2, D7 and D8 are rarely where a chemical supplier's 8D report fails. They depend on facts the supplier already holds: who worked the complaint, what the customer reported, which documents changed and when the report closed. D3 to D6 are different, because each needs evidence that has to be produced during the complaint, often within days: a count of suspect stock by lot, a retest of the retained sample, a cause chain that the retest supports, and success criteria written down before the corrective action runs. The sections that follow take those four disciplines in turn. Where a customer's own 8D form asks for something the table does not list, the customer's form decides what goes in the report.
What evidence does a chemical supplier need for 8D containment?
A chemical supplier needs evidence that every quantity of the suspect lot has been located, counted and held, at the plant, in transit and at the customer, plus the retained sample of that lot kept untouched for testing. A request to hold material, or a customer's agreement to a hold, is not containment until someone records that it was done.
Count suspect material by location. The SEW-EURODRIVE 8D form for suppliers asks for the suspect quantity and the quantity found not OK at the customer, in transport and at the supplier, separately (SEW-EURODRIVE, accessed 2026). For a chemical product, that means drums, totes or bulk tonnage by lot, the warehouse hold status, the shipments in transit and the customer's own stock, including material already charged to a process.
The same specification says immediate measures must be planned so that the material and method of the measure are free of faults, and that material approved for delivery after the measure must be marked in a way the customer can follow (SEW-EURODRIVE, accessed 2026). For a chemical supplier, a sorting or retest step needs a defined test, a named laboratory and a marking on released containers.
Protect the retained sample. Record where it is stored, who removed any portion and when. ASTM D4840-99(2018)e1, a guide written for water analysis, describes chain-of-custody procedures that document sample possession through collection, shipment, storage and analysis (ASTM D4840-99(2018)e1), and the same habit keeps a retained chemical sample usable as root cause evidence.
How does a chemical supplier prove a root cause in an 8D report?
A chemical supplier proves a root cause in an 8D report by testing the retained sample of the complained lot, comparing the retest with the CoA value within the test method's precision, and tracing the result to the batch record, the raw material lots and the process data. A cause that fits the symptom but has not been checked against the retained sample is a hypothesis.
Start with the measurement question. A customer's incoming test and the supplier's CoA are results from different laboratories, operators and equipment, which are reproducibility conditions as defined in ASTM terminology (NIST OSAC Lexicon, 2025). A difference between the two that falls inside the method's published reproducibility does not by itself show a lot defect (inference from the definition). For petroleum products and lubricants, ASTM D3244-24, a practice, gives a statistical way for a supplier and a receiver to combine independent results into an assigned test value and settle a dispute over specification conformance (ASTM D3244-24).
Then ask three questions, not one. The SEW-EURODRIVE specification requires the cause to be found along three 5 Why axes: why the fault occurred, why it was not discovered, and which system gap allowed it (SEW-EURODRIVE, accessed 2026). For a chemical lot, these are often a raw material or process deviation, a release test that cannot see the property the customer cares about, and a specification that never named that property.
The same specification calls a cause such as "lack of attention" unacceptable when a worker made the error, and asks for the reason for the erroneous action (SEW-EURODRIVE, accessed 2026).
How should a chemical supplier tie 8D corrective actions to the root cause?
A chemical supplier should tie each 8D corrective action to one line of the proven cause chain, so that every cause has an action and every action names the cause it removes. An action that cannot be traced to a cause is activity, and the customer's quality engineer will ask why it was chosen.
Write the cause chain from D4 as numbered lines: occurrence, non-detection and system gap. Then give each line its own action in D5 and D7. For an occurrence cause, the action changes the product or process: a raw material specification, a mixing time, a temperature limit. For a non-detection cause, the action changes what release testing can see, such as adding a test for the property the customer measured. For a system gap, the action changes the document that should have caught it, such as the product specification, the CoA template or the change control procedure.
Customers also judge whether the action is durable. The SEW-EURODRIVE specification states that measures such as training courses, the four-eyes principle and manual checking cannot meet its requirement for long-term measures, since human failure is not ruled out (SEW-EURODRIVE, accessed 2026). Not every customer is this strict, but a chemical supplier whose only corrective action is retraining should expect the 8D to come back.
Keep actions on the product separate from actions on the customer's use of it. If the cause chain shows that storage or dilution at the customer contributed, say so, and list those steps as recommendations to the customer rather than as the supplier's corrective action.
When should a chemical supplier fix 8D success criteria?
A chemical supplier should fix the 8D success criteria before the corrective action runs: which test or observation, what result counts as success, over how many lots or how long, and what result would reopen the 8D. Criteria written after the results are in can only describe what happened, not show that the action worked.
The SEW-EURODRIVE specification asks that the effectiveness of measures be defined and monitored, including the type of effectiveness observation and its duration (SEW-EURODRIVE, accessed 2026). For a chemical product, that usually means a named test method, a limit, a number of consecutive production lots and, where the complaint came from the field, a period of customer use without recurrence.
Specification disputes use the same discipline. ASTM D3244-24 requires the acceptance limit to be determined before testing begins, with the probability of acceptance agreed between both parties beforehand (ASTM D3244-24). That practice is designed for contracts covering petroleum products and lubricants, but the principle carries over to any 8D: agree the yardstick, then measure.
Write the criteria into D6 with their date, and keep the verification results apart from them. When the results arrive, compare them with the criteria as written, and report a failed criterion as a failed criterion. A customer who sees success criteria dated after the verification data will question every other date in the report.
What does an illustrative chemical supplier 8D report example look like?
An illustrative chemical supplier 8D report example shows D2 as a quantified problem statement with lot and CoA data, and D4 as a cause chain checked against the retained sample. Every product, lot number, value and limit below is illustrative and does not come from a real case. The table decides which causes the D4 retest supports (illustrative).
Item | Customer data | Supplier data |
|---|---|---|
D2 complaint | Separated layer in 2 of 12 drums, lot L-0412, metalworking fluid concentrate | Lot L-0412 shipped 2026-01-15, 12 drums |
D2 test value | pH 9.1 at incoming test | CoA pH 9.4; method reproducibility 0.4 (illustrative) |
D4 retained sample | Not applicable | Clear, single phase after 30 days at 20 °C (illustrative) |
D4 freeze-thaw retest | Drums stored outdoors, minimum about minus 8 °C (illustrative) | Retained sample separates after one freeze-thaw cycle (illustrative) |
The key row in this illustrative 8D example is the freeze-thaw retest: the retained sample is in specification at 20 °C but separates after freezing, which matches the customer's outdoor storage (illustrative). The pH difference of 0.3 is inside the method reproducibility of 0.4 (illustrative), so it is not evidence of a lot defect. The D4 cause chain then reads: occurrence, the product is not freeze-stable; non-detection, release testing has no freeze-thaw check; system gap, the storage temperature limit was not on the label or the technical data sheet. Each line gets its own action, and the D6 success criterion, set before the change, is no separation after three freeze-thaw cycles on three consecutive lots (illustrative).
Why do customers reject chemical suppliers' 8D reports?
Customers tend to reject chemical suppliers' 8D reports when containment is claimed but not counted, the root cause is asserted without a retained sample test, actions do not match causes, success criteria appear after the results, or the report misses the customer's deadlines. Each one leaves the quality engineer unable to close the complaint.
Containment is the first gap. A statement that "all stock is on hold" without a count by lot and location reads as an intention. The SEW-EURODRIVE form asks for suspect and not-OK quantities at the customer, in transport and at the supplier (SEW-EURODRIVE, accessed 2026).
Root cause is the second. Customers working under IATF 16949:2016 clause 10.2.3 need defined approaches for root cause analysis and verification of corrective action effectiveness (BSI, accessed 2026), so a cause such as "raw material variation" with no retest data will not satisfy them.
The third is a mismatch between causes and actions, especially when training is the only action. The fourth is success criteria that seem to follow the data rather than precede it.
The fifth is time. SEW-EURODRIVE expects a first response with D1 to D3 after 3 working days, and the response with D4 to D8 after the following 10 working days, with interim reports every 10 working days if that is not possible (SEW-EURODRIVE, accessed 2026). For returned defective parts, the 3 days run from their receipt at the supplier's site (SEW-EURODRIVE, accessed 2026). That is one customer's specification for its approved suppliers, but many customers set similar clocks, and a chemical supplier waiting on laboratory results should send the interim report rather than nothing.
What should a chemical supplier's 8D report checklist cover?
A chemical supplier's 8D report checklist should cover the lot data, containment counts, retained sample evidence, the three-part cause chain, the action for each cause and the success criteria fixed in advance, and each line should be true before the report goes to the customer.
D2 names the product, every lot number, quantities, dates, and the customer's test values beside the CoA values.
D3 gives suspect quantities by lot at the plant, in transit and at the customer, with hold records, not only hold requests.
The retained sample of each complained lot is located, protected and logged before any portion is tested.
Any comparison of customer and CoA results names the test method and states whether the difference is inside the method's reproducibility.
D4 states three causes: why it occurred, why it was not detected and which system gap allowed it.
Each cause is supported by a retest, a batch record or process data, and any inference is labeled as one.
Each action in D5 and D7 names the cause line it removes.
No cause is closed by training alone.
Recommendations for the customer's storage or use are listed apart from the supplier's corrective actions.
D6 success criteria, with test method, limit, number of lots or period, are dated before the action ran.
Verification results are reported against those criteria as written.
Revised specifications, release tests and procedures are listed with revision numbers.
The report meets the customer's response deadlines or an interim report is sent.
Lubinpla's troubleshooting task does not fill in the 8D form; working from your own documents and site history, it returns a cited troubleshooting report with the containment evidence, the root cause judgment, countermeasures tied to the cause chain and success criteria fixed before execution, and it does not judge whether executed countermeasures worked.