When a PAM Batch Underperforms: A Buyer’s Investigation and Claim Checklist

A lot has underperformed in use. Your first instinct is to call the supplier. Before you do, you need evidence — because without it, neither party can reproduce the observation, assign cause, or close corrective action. This checklist-based guide walks you through ten sequential steps: contain the affected lot without declaring fault, lock its full identity, write a precise unmet-requirement record, preserve chain-of-custody evidence, reconstruct the process timeline, classify possible causes using a neutral 5M+E framework, plan a matched-condition verification, assemble a supplier NCR and claim file, separate correction from corrective action, and verify effectiveness before you close the file. The same framework applies to a single underperforming lot and to a recurring drift across multiple shipments.

Step 1: Stop, contain and define the boundary

The first priority is protecting your operation and your evidence, not assigning blame. Quarantine the affected lot and any other lots from the same batch or production window, and stop dosing until you have defined the boundary of the problem. “Boundary” means: which containers, which delivery, which date range, and which process units received the material.

Three things must happen before any other step. First, physically segregate the affected containers from stock that is still in use — move them to a designated hold area, mark them clearly and restrict access. Do not continue dosing from the same lot while the investigation is open. Second, if safety is a concern, follow your applicable safety and disposal requirements; the investigation does not change those obligations. Third, secure the containers in a state that preserves labelling, seal condition and packaging evidence, because those will be needed later.

This step is containment, not a concession that the supplier made an error. Containment protects both parties: you preserve your operation and your evidence; the supplier retains the right to have the claim evaluated on actual evidence rather than on a description of consequences.

Step 2: Lock lot and product identity

Before any claim can be evaluated, the lot must be identified completely. A complaint that says only “product does not work” cannot be investigated. The reason is traceability: a PAM product identity is bound to the full grade designation, every suffix, the physical form, the issuer and the source family, and the version and test conditions under which it was characterised. Transferring a result or a specification from one grade to another, or from one batch document to another, is not valid without specific source evidence that they are comparable.

Complete this identity record from the physical container and the shipping documents, not from memory or from a previous lot’s paperwork. If any field cannot be confirmed from the available documents, record it as unconfirmed and note what document would resolve it.

Any inconsistency between the label, the COA and the packing list is itself a documented finding. Record the mismatch exactly, do not resolve it by assumption. For document cross-check guidance, including how to compare COA rows against TDS specifications and how to identify when a COA conclusion is inconsistent with its own result table, see the COA, TDS and SDS document review guide.

Step 3: Write the unmet requirement precisely

A well-documented nonconformity has three parts: the objective evidence, the requirement against which the observation is compared, and the statement of nonconformity. ISO/IAF Auditing Practices Group 2016 management-system guidance describes this structure and notes that the statement of nonconformity drives the cause analysis, correction and corrective action — so it must be precise, not a restatement of the evidence and not a conclusion about fault. This guide adapts that record discipline to a buyer investigation of an underperforming lot: evidence first, then the requirement, then the statement.

The requirement is whatever the lot was supposed to meet: a parameter in the purchase order, a value in the approved TDS specification, a buyer-defined acceptance criterion from the qualification record, or a contractual performance metric. If no written requirement exists and you are working from informal expectations, that is itself a finding worth recording — it is harder to defend a claim without a written requirement, and harder to request corrective action without a defined reference.

The observed result must be specific: the measured value, the method used to measure it, the units, the conditions under which the measurement was made (concentration, temperature, feed, equipment), and the date. “Product did not flocculate well” is not a result. A complete result statement uses your own plant data in the form: “[Measured parameter] = [your recorded value] [unit], against [your acceptance criterion] from [source document], measured by [your test method] at [your dose, feed conditions and temperature], on [date from your records]”. The source document is your purchase order, TDS specification, acceptance criterion from the qualification record, or contractual performance metric — not a number supplied in this article, which does not know your application or requirements.

Retained and current samples need clear identity, conditions and methods before results can be compared. A laboratory observation without a defined method, unit and condition cannot support a reproducible claim.

Step 4: Preserve evidence and maintain chain of custody

Evidence that is not preserved cannot support a claim. Evidence that is handled without a custody record may be disputed. Both are avoidable.

The evidence set for a disputed lot includes: the physical containers (opened and unopened, with original markings intact), any retained incoming sample that was secured before the bulk was put into service, the label photographs taken on receipt, the COA and TDS filed for this specific lot, the delivery documents, and the process records from the period when the material was in use. If you qualified this grade through a plant trial and retain a sample from that qualification, that retained sample is also evidence — it allows a controlled comparison under matched conditions.

A chain of custody record is a log that tracks who had access to the evidence, when, and what they did with it. The log must support traceability; the required form is governed by applicable contract, site procedure and applicable law. It needs to be complete enough that a supplier representative reviewing the investigation can confirm that the retained sample and the containers were stored under defined conditions and not substituted or contaminated between receipt and the time any comparison test is run.

Seal condition, handling history and container identity belong in the evidence chain for a disputed lot. Record whether seals were intact on receipt and photograph label and seal condition before any investigation handling begins.

Retain the following for every disputed lot: identifiable unopened material from the lot in the quantity required by the buyer’s documented site or hold procedure and packaging format, with actual quantity, state and location recorded; the COA, TDS and SDS as-received; delivery documentation; label photographs showing lot number, grade, seal and packaging condition; process records including dosing logs, solution preparation records and performance measurement records for the period the lot was in use; and the retained incoming sample if one was secured at receipt. If your incoming QC protocol did not include retaining a sample, record that explicitly — it does not prevent the investigation, but it means a direct sample comparison against your qualification reference is not available. For guidance on incoming-lot release including retained-sample practice, see the sample-to-bulk qualification guide.

The custody/access log below is the record that tracks who had contact with the evidence, when, and what was done. Maintain an entry for every transfer, inspection or action from the point of containment through the close of the investigation.

Step 5: Reconstruct the process timeline and baseline

One investigation gap is assuming the lot is the cause before the process variables have been reviewed. A buyer may attribute a poor result to the polymer while feed, dilution, maturation, storage, equipment settings, operator steps or measurement method also changed. A reconstructed timeline that captures all variables from before and after the event provides the evidence needed to classify each factor.

The timeline covers: when the lot was delivered and put into service; what the feed characterisation was for the process unit that week (total solids, pH, temperature, feed rate — or the equivalent parameter set for your application); how solution was prepared (concentration, water quality, mixing time, maturation time, and any deviation from normal); what dose was applied; how performance was measured and when; and what other changes occurred in the process during the same period — equipment maintenance, upstream process changes, other chemicals added, operator changes, seasonal temperature shift.

Write the timeline out in chronological order. Include entries from the period before the event, not only from the event date. A performance drift that started before the new lot arrived is a different investigation from one that started on the day the lot was first dissolved.

The goal is a baseline: a documented description of what the process was doing before the lot in question arrived, so that any comparison test later can be interpreted against a defined reference rather than against a recollection. If the incoming lot is dissolved and dosed under conditions that differ from the baseline, a different result does not prove the lot is nonconforming; it proves the comparison conditions were different.

Stable recorded conditions across the period in question may raise the priority of the material-side hypothesis, but do not confirm it; additional evidence from the verification and cause-analysis steps is needed before a material cause can be classified as confirmed. Where simultaneous process changes occurred during the same period, those changes remain hypotheses to classify in the 5M+E step below — they neither exonerate the material nor confirm it.

Step 6: Build a neutral 5M+E cause tree

Before any cause can be described as confirmed, it must be evidence-supported. ISO/IAF Auditing Practices Group 2016 management-system guidance recommends examining a list of possible factors and categorising each as “not able to confirm”, “possible”, or “confirmed” in terms of potential impact on the root cause. This classification should be iterated: first pass identifies possible factors; further work may classify a factor as “confirmed”, retain it as “possible”, or leave the determination inconclusive where the evidence does not resolve it. Do not stop at the first plausible factor. For a PAM underperformance investigation, the 5M+E framework — people, machine, material, method, measurement, environment — is adapted below to buyer process variables.

PAM-specific process variables — incorrect solution concentration, inadequate maturation time, feed pH shift, seasonal temperature drop, upstream coagulant change, equipment wear — are hypotheses until the reconstructed timeline and comparison evidence support their promotion. A residual hypothesis that cannot be confirmed should be recorded as “possible, not confirmed”, not dropped from the record.

This matrix does not replace the investigation; it structures it. Fill in what you observe and can document; leave fields blank or “cannot confirm” rather than speculating. For dissolution issues specifically, the lump, fisheye and dissolution troubleshooting guide covers the material-side dissolution failure modes. For overdosing-driven performance reversal, see the overdosing detection and fix guide. Both are separate from this investigation framework; link to them from the relevant cells rather than repeating their content here.

Step 7: Plan a matched-condition verification

A retained-sample comparison is an informative test when a lot is disputed: dissolve the lot sample and a reference sample (retained incoming from a previous accepted delivery, or retained qualification sample) under identical conditions and compare the results. Comparability depends on matched, recorded conditions: same concentration, same make-up water source and quality, same mixing procedure, same maturation time, same test method, same feed if a bench application test is run. An unmatched comparison has limited or non-direct comparability and must be labelled with its differences; it is not a substitute for a matched-condition test.

Before running a comparison, define the acceptance plan: what result would support “lot meets requirement”, what result would support “lot does not meet requirement”, and what result would be inconclusive. Define these before you see the comparison result, not after. A post-hoc acceptance rule is not a controlled verification.

This plan also applies when the supplier proposes a verification. If the supplier runs a re-test on retained production material or on a sister lot from the same batch, and the conditions under which they test are not disclosed or are not matched to the buyer’s conditions, the result is not directly comparable to the buyer’s observation. Ask for the test protocol and the preparation conditions before accepting a supplier re-test result as a resolution.

If a comparison test is not feasible because no retained incoming sample exists, or because the lot material is exhausted, document that constraint clearly. It does not end the investigation, but it changes the evidence basis: the claim rests on the process records, the COA cross-check and the 5M+E analysis rather than on a direct sample comparison.

Step 8: Assemble the supplier NCR and claim file

Once Steps 1–7 are complete, you have the evidence to approach the supplier. If your contract sets notice deadlines for quality disputes, observe those deadlines while completing the file; initial notice to the supplier can precede the complete investigation record, as long as the evidence file follows promptly. A claim file that arrives complete and organised invites a technical response.

The claim file contains the evidence, the nonconformity statement, a technical response request and — separately — a commercial discussion request. Keep the technical and commercial parts distinct so each can proceed on its own timeline and on its own evidence basis.

ISO/IAF Auditing Practices Group 2016 management-system guidance notes that a well-structured NCR record facilitates progress reviews and evidences corrective action and its effectiveness. This guide adapts that record discipline to a buyer investigation: the more complete and structured your file, the more useful the supplier’s response is likely to be.

Step 9: Choose disposition and separate correction from corrective action

Disposition is a decision about the held material. Corrective action is a change to prevent recurrence. These are different actions with different owners and different timelines; keeping them separate supports a clearer investigation record. ISO/IAF Auditing Practices Group 2016 management-system guidance defines correction as action to eliminate a detected nonconformity, and corrective action as action to eliminate the cause of a nonconformity to prevent recurrence. Correction can proceed independently; corrective action should be tied to the best-supported cause analysis available.

Disposition options for a held lot include: use-as-is (when investigation shows the lot meets requirements and the initial observation is explained by process variables); rework or retest under defined conditions; hold pending investigation outcome; return to supplier; or dispose of the material subject to applicable safety and disposal rules. The applicable rules for PAM waste disposal depend on your jurisdiction and the specific grade; consult the SDS and your site’s environmental procedures. The investigation does not suspend those obligations.

Correction — the immediate action on this lot — might be restoring the process to its pre-event baseline, adjusting dose under a controlled plan, or accepting the loss and ordering a new lot. Corrective action — preventing the same lot-level problem in future — should be tied to the best-supported cause available from the analysis. Where the investigation outcome remains possible or inconclusive, record the uncertainty, the controls in place and the monitoring plan without presenting an unconfirmed hypothesis as a confirmed root cause; track the corrective action as open until effectiveness evidence closes it.

Where cause analysis, contract terms and the record of which party controlled each contributing factor support assigning technical actions or commercial responsibility, assign them accordingly using that evidence. The supplier’s technical response should address the confirmed material-side findings, including any root-cause explanation and prevention commitment relevant to those findings. Where causes are shared, or where a material-side cause remains possible but not confirmed, record the open hypotheses and the monitoring plan; do not assign commercial responsibility ahead of the evidence. If the investigation is inconclusive, document the open hypotheses and the monitoring plan before partially closing the file.

Step 10: Verify effectiveness and close the file

A corrective action that is implemented but not verified may or may not have worked. ISO/IAF Auditing Practices Group 2016 management-system guidance states that closeout needs objective evidence that the described corrective action has been fully implemented and is effective in preventing the nonconformity from recurring. This guide adapts that record discipline to a buyer NCR: closing the file requires evidence of implementation and evidence of effectiveness, not only a statement that the action was taken.

Implementation evidence is a record that the specific action was carried out: a training record, a revised procedure, a supplier confirmation of production change, a revised incoming-QC protocol, a new retained-sample requirement. Implementation evidence shows what was done, not whether it worked.

Effectiveness evidence is a subsequent record showing the nonconformity did not recur under conditions where it would have been expected to: the next delivery of the same grade was checked under the incoming plan and met requirements; the process performance metric was back within the acceptance range for a monitored period proportionate to the risk and the frequency of the affected activity; the 5M+E factor that was confirmed as the cause has been controlled and there is no recurrence signal. Define the effectiveness check criteria and monitoring window before the corrective action is closed.

Product identity and batch evidence remain bound to the specific lot, grade and source family for which they were generated. Do not generalise a result from the investigation lot to future lots or to other grades without separate evidence for each. A supplier corrective action that resolves the cause for one production run does not automatically certify future lots.

A file that is closed without evidence of effectiveness has documented actions, not results. A file that is closed with effectiveness evidence is the record you need if the same problem recurs: you can show what was done, when, by whom, and that it worked for a defined period. If the problem recurs after a closed NCR, re-open the file, revisit the cause analysis and consider whether the corrective action addressed the current or best-supported cause determination or a symptom of it.

For guidance on supplier qualification before a purchase is placed — the controls that, if in place, can reduce the investigation load when a lot is disputed — see the supplier audit and qualification checklist.

ChinaPAM can receive and review the complete evidence file — full grade and suffix, lot number, label photographs, COA, TDS and SDS for the lot, observed result with method, units and conditions, process timeline and retained-sample status. Where the evidence supports it, we can discuss a controlled comparison and quote an available verified supply route. Send the package to the contact form or through your existing order channel.

Frequently asked questions

Do I need to complete all ten steps before contacting the supplier?

Applicable contract notice deadlines govern prompt notice; check your purchase agreement for any quality-dispute notice window. For the technical exchange, completing Steps 1–4 before sending a detailed claim — contain the lot, lock its identity, write the unmet requirement, and secure your evidence — makes the exchange more specific and reviewable for both parties. Steps 5–7 strengthen the file and are worth doing before Step 8 if your process records allow it — they help separate material-side from process-side causes and demonstrate that the investigation was disciplined. Steps 9 and 10 follow from the supplier response.

What if I have no retained incoming sample from this lot?

A missing retained incoming sample changes the verification options but does not end the investigation. Document explicitly that no retained sample was secured. The evidence base then rests on the process records, the COA comparison, the 5M+E analysis and any sealed containers that remain. If containers remain, preserve identifiable unopened material per the documented site or hold procedure, recording quantity, state and location before further handling. If the lot is entirely consumed, the claim rests on the process and document record alone; a direct sample comparison is not available for this lot. For future deliveries, see the sample-to-bulk qualification guide for incoming-sample retention practice.

The supplier says their own re-test passed. Does that close the investigation?

A supplier re-test result on retained production material or a sister lot is one data point. It does not by itself resolve the buyer’s observation, because the test conditions, preparation method, feed and equipment used by the supplier may differ from the buyer’s process conditions. Ask the supplier to provide the test protocol, preparation conditions, and the identity of the material tested (lot number, whether it is from the same production batch as the disputed lot or a different one). Then assess whether the supplier’s result is comparable to the buyer’s observation. If conditions differ, the results are not directly comparable until the condition differences are evaluated and the observed gap is explained.

How long should the effectiveness monitoring period be?

Define the monitoring period and the recurrence criteria before you close the corrective action, based on the nature of the cause and the frequency of the affected activity. The period should cover enough deliveries or production cycles to show the corrective action holds under normal operating variability. There is no universal answer; the period should be proportionate to the risk and the confidence in the current cause determination. Document the chosen window and criteria in the corrective-action record so the effectiveness check can be assessed against a pre-defined standard rather than a post-hoc judgement.

Sources

• ISO 9001 Auditing Practices Group / IAF — Nonconformity: Documenting (13 January 2016). PDF via ISO committee site. Supports: a well-documented nonconformity has three parts — objective evidence, the unmet requirement, and the statement of nonconformity; evidence must be sufficiently detailed for another person to confirm the observation; systematic NCR records support traceability and effectiveness evidence. Limits: educational management-system guidance published by the ISO 9001 Auditing Practices Group and IAF; carries an explicit non-endorsement disclaimer; not a PAM product standard, a PAM test method, a contract remedy, a ChinaPAM certificate, or a universal acceptance threshold.
• ISO 9001 Auditing Practices Group / IAF — Nonconformity: Review and Closing (13 January 2016). PDF via ISO committee site. Supports: correction eliminates a detected nonconformity; corrective action eliminates its cause to prevent recurrence; cause factors should be classified as not able to confirm / possible / confirmed before promotion to root cause; closeout requires objective evidence of implementation and effectiveness. Limits: same educational guidance scope as above; not a normative PAM product standard or contractual entitlement.