A useful quality concern describes an observable discrepancy and connects it to identifiable material. It does not need to begin with a diagnosis of the cause. Clear records make it easier to distinguish a product issue, a document mismatch and a question about how evidence was interpreted.
Describe the concern before assigning a cause
ICH Q10 includes complaints among the inputs to structured quality investigation and corrective action. Its pharmaceutical quality-system model illustrates why a concern can be useful evidence rather than simply a customer-service exchange.ICH Q10 — Pharmaceutical Quality System (opens in a new tab)
An original example is a vial marked lot B-14 supplied with a certificate for B-13. The observable concern is the mismatch. It does not establish that the material is counterfeit, chemically incorrect or untested.
Another concern might be a visible change in material or a discrepancy in an independently obtained analytical result. State what was observed, when and on which unit, while distinguishing direct observation from an interpretation.
Build a compact evidence record
| Record | Useful detail |
|---|---|
| Product and lot | Exact label wording, size and readable lot identifier |
| Order context | Order reference and receipt date |
| Observation | What differs from the stated expectation |
| Documents | Complete report, identifier and version |
| Material history | Known opening, handling and storage events |
| Requested outcome | The explanation or investigation needed |
Keep original photographs and documents. An annotated copy can point to a detail, but retaining the original avoids losing context or confusing added marks with source information.
For an analytical discrepancy, include the measured property, unit, sample basis and method information available. Two different percentages may not be directly comparable, so preserve the reports rather than sending only the selected numbers.
Record unknown history as unknown. Guessing a storage condition or reconstructing an exact time from uncertain memory can make the investigation less reliable.
Ask for an outcome that addresses the discrepancy
For the fictional lot mismatch, a focused request would ask which certificate applies to the supplied B-14 vial and what record establishes that connection. Asking only whether the product is good leaves the actual mismatch unanswered.
For an analytical difference, ask whether the supplier will review the two sample and method bases and explain any proposed follow-up. Repeating a test may be useful, but its purpose and material should be clear.
Keep the request proportionate to the evidence. A photograph of a label can establish what the label says; it cannot measure purity. A laboratory result can establish a finding for its sample; it cannot automatically identify the history of every unit in an order.
If physical material may need examination, preserve its existing identity and records while agreeing suitable arrangements with the recipient. Do not alter or combine material simply to make the concern easier to demonstrate.
Keep investigation findings separate from the remedy
A refund or replacement can resolve the commercial part of a concern without explaining its technical cause. Conversely, a technical explanation does not by itself tell the customer how the order issue will be handled.
Retain the response, supporting document versions and any investigation reference together. If the explanation introduces a new result, check which material was tested and how it relates to the original concern.
A useful closure records what was established, what action followed and what remains unresolved. This article does not promise a particular supplier response time, testing service or returns outcome; those depend on the actual arrangements.
Sources and further detail
- ICH Q10 — Pharmaceutical Quality System (opens in a new tab)
Section 3.2.2 read for complaints as investigation inputs. The evidence checklist and B-13/B-14 example are original practical writing guidance; no message sent or Novum policy invented.
Sources checked 19 September 2026. Worked examples are illustrative unless a supplied report is explicitly identified. This article has not undergone independent scientific peer review.