Start with what a COA can and cannot establish

A certificate of analysis is a controlled summary of results reported for a material or lot. It can help a reviewer understand which attributes were examined, which methods were named, which results were reported, and whether those results were compared with a stated specification. It is not automatically proof that the document is authentic, that the sampled material is the same material being offered, or that the product is suitable for a particular project.

Treat the COA as one link in a traceability chain. The product page, quotation, container label, packing record, catalog number, lot or batch number, and COA should refer to the same item. A high percentage, a signature image, a QR code, or a polished template does not replace that match. If the chain cannot be completed, record the uncertainty instead of treating the document as confirmation.

Match the document to the exact material

Read the identification block before reviewing any result. Compare the formal material name, stated form, catalog number when provided, lot or batch number, document number, issue or test date, and revision. The material form matters because a salt, hydrate, counterion, modified sequence, or other defined form may have different reference information from a similarly named entry.

The lot number is especially important. A generic example COA can illustrate a document format, but it is not lot-specific evidence for material that will be supplied later. If a supplier provides only a sample document, label it SAMPLE or REPRESENTATIVE in your records and request the actual lot-linked document at the appropriate stage. Never remove that distinction when publishing or sharing the file.

Separate identity, purity, assay, and other attributes

Different tests answer different questions. An identity result asks whether the observed characteristics are consistent with the stated material. A chromatographic purity result describes the relative signal distribution under the stated method and conditions. An assay may estimate how much of a defined component is present relative to a standard or calculation. Water, residual solvent, counterion, or other measurements may be reported separately.

These values should not be collapsed into a single quality number. For example, a chromatographic area percentage is not automatically the same as mass fraction, identity, assay, sterility, or suitability. Read the method name, units, calculation basis, and specification attached to each result. If the basis is not stated, ask what the number represents rather than supplying your own interpretation.

Read method, specification, result, and disposition together

A useful result row normally connects four elements: the attribute being evaluated, the method or controlled procedure, the acceptance specification, and the observed result. Units should be explicit where they apply. Terms such as conforms or passes are meaningful only when the corresponding specification and method are identifiable.

Check whether the result is numerical, qualitative, calculated, or simply a disposition. Check whether rounding could affect the comparison and whether a less-than sign represents a reporting limit rather than zero. If a method code is listed without an accessible explanation, record the code exactly and ask the document owner what the method measures. Do not invent an analytical procedure from an abbreviation.

Look for document control and traceability

Document control helps a reviewer understand where a COA came from and which version is current. Useful fields can include the issuing organization, document or certificate number, revision, page count, approval role, issue date, lot number, and a way to verify the document through a known supplier channel. Their presence improves traceability, but none of them should be fabricated when absent.

Flag mismatched names, altered fonts around key values, missing pages, conflicting dates, a lot number that differs from the label, specifications without results, results without units, or claims of independent testing without the laboratory identity and report connection needed to support them. A flag is a request for clarification, not an accusation. Preserve the original file and the supplier's response as part of the review record.

Use a repeatable review record

Create a simple review table with columns for source file, date received, product name, catalog number, lot number, material form, reported tests, methods, specifications, results, units, document owner, and open questions. Mark each field VERIFIED, NOT PROVIDED, CONFLICT, or REVIEW REQUIRED. This makes gaps visible without turning assumptions into facts.

Before relying on the COA, confirm that the document belongs to the intended item and lot, that every quoted result is represented accurately, and that the supplier can explain unresolved fields. Keep the COA separate from medical, regulatory, or end-use conclusions it does not establish. The final decision about procurement or use belongs to the responsible organization and its applicable technical, quality, and legal review.

Scope and limits
  • Educational document-review content only; no human-use, dosing, treatment, efficacy, or weight-loss content.
  • Does not state that a COA proves authenticity, suitability, legality, independent testing, purity, or certification.
  • Any product-specific result, storage statement, specification, or lot claim must come from a verified source document before publication.

Further reading: primary sources

These sources provide background on certificate contents and interpreting analytical results. Each has its own scope; neither source reviews or endorses this article, NorthVerity Bio, or any listed product.