Buying and sourcing

Reading a certificate of analysis: what testing does and does not prove

A certificate of analysis is a report about a sample. It is not a report about a product, a batch, or a seller. Most of the confusion around these documents comes from collapsing those four things into one.

What a certificate of analysis is

Read strictly, a certificate of analysis makes a statement of one shape: on this date, using this method, this laboratory measured these values in the material submitted under this identifier. That is the whole of it.

Almost everything a buyer wants to know sits outside that sentence. Whether the tested material came from the batch that produced your vial, what happened to your vial afterwards, and whether the document in front of you is the one the laboratory actually issued are all questions the certificate cannot answer about itself.

The anatomy of the document

Most of a certificate is identification rather than measurement. Before reading a single number, check whether the document identifies itself completely.

• The issuing laboratory, as a legal name and address rather than only a logo.
• Any accreditation reference, with the scope and certificate number that can be checked against the accreditation body's own register.
• A report date and a sample receipt date, which are not the same date.
• A sample identifier assigned by the laboratory, distinct from whatever name the submitter supplied.
• The batch or lot number attached to the sample by whoever submitted it.
• The analytical method behind each result, named specifically enough to be looked up.
• Acceptance criteria or specification limits, where the laboratory was asked to judge a result against them.
• A signature or authorising analyst, with a stated role.¹

The one structural question

The most important field on the page is the lot number, and the only thing worth comparing it against is the vial in front of you. If the certificate carries no lot, or carries a lot that does not appear on your vial, the document describes different material and the rest of the reading exercise is decoration.

The second structural question is who chose and submitted the sample. A laboratory tests what arrives. If a seller selected the vial, packaged it, and posted it, the result describes the seller's chosen sample under the seller's chain of custody. That differs from a sample drawn by an independent party from the same lot as the unit you received, and certificates rarely record which situation applies.

What accreditation means

ISO/IEC 17025 sets out general requirements for the competence, impartiality, and consistent operation of testing and calibration laboratories. Accreditation to that standard is granted by an accreditation body, for a defined scope of tests, and is recorded in that body's register.¹

Two limits follow. Accreditation is scoped, so a laboratory can hold it for tests unrelated to the one on your certificate. And accreditation is a status held by a laboratory, not a property of a PDF. A logo, a certificate number rendered as an image, or the phrase “accredited laboratory” in a footer is a claim; the register is the evidence.¹

Reading the method section

You do not need to be an analytical chemist to read a method line honestly. You need to hold one idea steadily: a result is a number produced by a procedure, and changing the procedure changes the number.

What a purity figure means

Chromatographic purity is usually reported as the area of the main peak as a percentage of total detected peak area. That is a percentage of what the detector saw, under one column, one gradient, one wavelength, one run time. Substances that do not respond to that detector, do not elute within that run, or co-elute with the main peak never enter the denominator. Two laboratories can report different purities for the same material without either being wrong.

The peptide-impurity literature describes why a single number hides so much: related substances in peptide preparations include deletion and insertion sequences, truncated and elongated chains, diastereomers, oxidation and deamidation products, residues of incompletely removed protecting groups, and aggregates. These are distinct classes with distinct behaviour, and one percentage does not separate them.³

What identity testing means

Mass spectrometry on this kind of report generally confirms that a measured mass matches an expected mass. That is useful and bounded. A matching mass is consistent with the intended sequence; it does not by itself exclude every isomer, stereochemical variant, or same-mass substitution, and it carries no information about how much of that material is present.

What is not on the certificate

Sterility, bacterial endotoxin, residual solvents, water content, counterion identity and content, and elemental impurities are separate tests with separate methods. If they are not listed with a result, they were not reported. An absent test is not a passed test, and a certificate that covers purity and identity alone has not addressed them.

One sample, one moment

A certificate describes the material that was tested, on the day it was tested. It cannot describe the other vials filled from the same batch, because they were not tested. It cannot describe what happened to your vial afterwards, including shipping temperature, time in a warehouse, light exposure, or repackaging. And it cannot vouch for itself: a PDF is trivially editable, and a certificate that has been altered, reused across batches, or copied from another supplier looks exactly like one that has not.

That makes chain of custody, rather than the numbers, the load-bearing part of the document, and chain of custody is the part a buyer usually cannot verify. Asking the named laboratory to confirm that it issued a report under that identifier is the only ordinary check available, and many certificates in circulation do not name a laboratory that answers.

The regulatory frame

Two regulatory facts sit behind all of this. The first is that unapproved and counterfeit GLP-1 products are a documented supply-chain problem rather than a hypothetical one: the FDA has published concerns about unapproved GLP-1 drugs marketed for weight loss, and separately warned consumers about counterfeit semaglutide identified in the U.S. drug supply chain.²⁶

The second is that “validated method” has a defined meaning. Regulatory guidance on validation of analytical procedures sets out what establishing a method's specificity, accuracy, precision, range, and robustness involves for a stated purpose. Consumer-facing certificates frequently name a technique without any indication that the procedure was validated for the material being tested, which is a weaker claim wearing similar language.⁵

A separate frame applies to sport. Anti-doping rules operate on substances and methods, independently of purity, provenance, or paperwork. A substance on the World Anti-Doping Agency's Prohibited List remains prohibited for an athlete under those rules whatever a certificate reports, and a favourable analysis creates no exemption.⁴

Where independent verification fits

Commissioning your own test changes one specific thing: it produces a measurement whose chain of custody starts with you. You chose the unit, you submitted it, and the result describes that unit rather than a sample selected by a seller. That closes one failure mode, the document that does not correspond to the material you hold.

It closes none of the others. An independently commissioned report still measures one sample at one point in time, using methods whose scope and validation you would still have to read, and it still says nothing about sterility, legality, or whether any substance is appropriate for any person. It is a second, independently commissioned measurement, not a clearance.

Reading a certificate well ends in a narrower place than it starts. This document, if genuine, reports these values for one sample that someone submitted on this date. Everything past that sentence is inference, and most of it is unsupported.