Reading a certificate of analysis

What a COA should establish

A certificate of analysis, or COA, is meaningful only to the extent that it connects a defined material to specific analytical work. The document is not itself proof of quality; it is a structured record of tests, methods, results and traceability. A well-constructed COA allows a reader to answer four basic questions: what material was tested, which batch was represented, what was actually measured, and who performed or authorised the analysis.

Batch linkage and identity

The first requirement is unambiguous identity. The certificate should state the material name and, where relevant, sequence, chemical form, strength or nominal fill. The batch or lot number is critical because analytical results apply to the sample tested, not indefinitely to every future production run carrying the same product name. If a physical vial cannot be matched to the lot stated on the certificate, the evidence chain is incomplete. Ideally, the supplier can move in both directions: from vial to batch record and COA, and from the batch record to the units released under that identifier.

Tests and methods

Next, examine the tests and methods. A peptide COA may include HPLC or UHPLC for chromatographic purity, mass spectrometry for identity, water content, residual solvent, counter-ion analysis, appearance, fill quantity or other specifications. The method should be named with enough detail to understand what the result represents. A bare statement such as “purity: 99%” is weaker than “HPLC-UV main-peak area: 99.2%” because the latter identifies the measurement basis. The date of analysis is also important, particularly for stability-sensitive materials.

Identity is not purity

Identity and purity should not be conflated. A chromatographic purity result indicates the relative distribution of detected peaks under the method. A mass-spectrometric result indicates whether the measured mass is consistent with the expected molecule. Neither necessarily establishes absolute peptide content by weight. If the document reports multiple attributes, each should have its own method and result rather than being implied by a single headline number.

Who performed the analysis

The laboratory matters. If analysis was conducted by an independent third-party laboratory, that laboratory should be named and the relationship described accurately. Where available, laboratory identifiers, report numbers, task numbers, signatures, verification links or digital verification keys improve traceability. They also make it easier to distinguish an original laboratory report from a supplier-created summary. A supplier may legitimately compile results, but it should not imply that third-party work was performed internally.

Red flags and omissions

Readers should also look for omissions and inconsistencies. A COA without a batch number, a date that predates the stated batch, a product description that does not match the label, missing method details, recycled chromatograms or results that appear across multiple unrelated lots are all reasons

for further investigation. Typographical errors do not necessarily invalidate analysis, but unexplained mismatches weaken confidence in document control.

The COA as part of a quality system

A strong COA is therefore best read as part of a quality system. It should link to the physical material, reflect tests actually performed on that lot, preserve raw or source records, and use scientifically accurate terminology. The key question is not “does a certificate exist?” but “can this certificate be traced to this batch, and do the reported methods support the claims being made?” That approach turns the COA from a marketing artefact into a technical record.