How to verify a certificate of analysis

A certificate of analysis (COA) is the single document that separates a supplier making a claim from a supplier producing evidence. It is also the document most often faked, misread, or quietly substituted for something weaker. This guide explains what a legitimate COA contains, how to read one, and what should make you walk away — written to apply to any supplier, including us.

What a COA is — and what it is not

A COA is a record of what a laboratory measured on a specific manufactured lot. It is not a certification of the company, not a licence, and not a statement that a compound is safe or effective for any use. A supplier that says “we are third-party tested” without producing a lot-specific document has told you nothing verifiable.

The critical word is lot. Material is made in batches, and each batch is chemically distinct. A COA for lot A says nothing about the vial in your hand if that vial came from lot B. Any supplier who shows one certificate for a product they have been selling for two years is showing you a marketing asset, not a record.

The six things a real COA states

1. The lot number — and it must match the number printed on the vial you received. If you cannot connect the paper to the physical object, the paper is decorative.
2. Identity — confirmation that the material is the compound named, usually by mass spectrometry. Identity and purity are different questions; see below.
3. Purity, with the method named — typically HPLC, reported as a percentage of the main chromatographic peak.
4. The date of testing — a COA with no date cannot be placed relative to the manufacture date.
5. The testing laboratory — an identifiable lab, not “independent third party” in the abstract.
6. A signature or authorised release — a named person or role accepting responsibility for the result.

Identity versus purity — the most common misreading

These answer different questions and a document can pass one while failing the other.

Purity asks: of everything in this vial, what fraction is the main component? Identity asks: is that main component actually the molecule on the label?

This distinction matters because a vial can be 99% pure and 100% wrong. If a lot were filled with a different peptide entirely, an HPLC purity assay could still return a clean 99% main peak — the instrument reports that one substance dominates, not which substance it is. Only an identity test, typically mass spectrometry against a reference standard, closes that gap. A COA reporting purity but not identity is incomplete, and it is the most common way an otherwise real-looking document fails.

What HPLC purity does and does not measure

High-performance liquid chromatography separates a sample into components and measures the relative size of each. “99% HPLC purity” means the main peak accounts for 99% of what the detector saw under those conditions.

It does not measure:

• Peptide content. Lyophilised peptide contains residual water and counterion salt. A vial can be 99% pure by HPLC and still contain meaningfully less peptide by mass than the label states. Net peptide content is a separate assay.
• Anything the detector cannot see. Residual solvents, heavy metals and endotoxin do not appear as HPLC peaks. They require their own methods — ICP-MS for elemental impurities, USP <85> for endotoxin.
• Sterility. A sterile product requires a sterility test (USP <71>). Purity says nothing about it.

So “99% pure” is a real and useful number, and it is also a narrower claim than most buyers read it as.

Red flags

• No lot number, or a lot number that does not match the vial. The defect that makes everything else moot.
• One COA covering the whole product line. Certificates are per-lot by definition.
• A flat image with no provenance. A JPEG of a certificate can be edited in a browser. Prefer a document you reached from the supplier’s own domain, ideally resolved from the lot number itself.
• An unnamed laboratory. “Third-party tested” with no testable referent.
• Purity quoted with no method. A percentage with no assay behind it is a number, not a measurement.
• Suspiciously round or identical results. Real assays scatter. Every lot landing on exactly 99.0% is a sign of a template rather than a measurement.
• A date that precedes the manufacturing date, or no date at all.

How to verify independently

Reading a COA and verifying it are different acts. To verify:

1. Start from the vial, not the website. Take the lot number off the physical label and look it up, rather than browsing to a certificate the supplier chose to show you.
2. Confirm the document resolves on the supplier’s domain at a URL tied to that lot.
3. Check the named laboratory exists and performs the assays claimed.
4. Compare the reported method to the claim. Sterility asserted with no USP <71> result is an unsupported claim.
5. Send your own sample. The strongest form of verification is one the supplier did not control. Independent testing services accept customer-submitted vials, and a supplier confident in its material has no reason to discourage this.

How Merit handles this

We publish the certificate for every lot before it is sold, and each lot number resolves to its own page on this domain — the same lookup described above. The vial label carries a code that leads to that record. Where a lot has been tested but the signed artifact cannot be published, we say so on the page rather than omitting the lot.

We state this here because a guide that tells you to demand lot-level evidence is only worth reading from a supplier that produces it. Apply the checks above to us as readily as to anyone else.

All materials discussed are supplied for laboratory and scientific research use only — not for human or veterinary use, and not FDA-approved. Nothing here is guidance on preparing, administering, or using any compound.

Frequently asked

Does a COA mean a compound is safe?

No. A COA reports what a laboratory measured on a lot — identity, purity, and whichever additional attributes were tested. It makes no statement about safety or suitability for any use, and research compounds are not approved for human use.

Is 99% purity better than 98%?

On the same assay, yes, marginally — but the more important question is whether identity was confirmed and which impurities make up the remainder. A well-characterised 98% lot with confirmed identity and a full impurity profile is a better-understood material than a 99% figure standing alone.

Can a certificate of analysis be faked?

Yes, and flat image files are trivial to alter. This is why lot-level resolution on the supplier’s own domain matters more than the visual appearance of the document, and why customer-submitted independent testing is the strongest check available.

What is the difference between a COA and third-party testing?

A COA is the document. “Third-party testing” describes who ran the assay — a laboratory independent of the manufacturer. A COA produced by an in-house lab is still a COA; it simply carries less independence. The certificate should name the laboratory so you can tell which you are looking at.

Why do some lots have no published certificate?

Sometimes a lot is tested but the signed document contains information a supplier cannot publish. The honest handling is to disclose that the lot exists and that the artifact is withheld. Quietly omitting the lot is not.