How to read a peptide certificate of analysis

Almost every peptide vendor now says "lab tested". Far fewer will show you the document, and fewer still produce one that would survive ten minutes with someone who reads certificates for a living. This is a practical guide to being that person — what a certificate of analysis (CoA) is for, what each field is actually telling you, and the handful of checks that separate a real one from a decorative one.

What a certificate of analysis is — and isn't

A CoA is a record of measurements made on a specific batch of material, written up against that batch's identifier. It answers a narrow question: when this batch was tested, what did the instruments find?

It is not a licence, an approval, a safety assessment or a promise about the batch you will receive next month. Anyone who reads it as those things is reading too much into it. Anyone who won't show it to you is asking you to read nothing at all.

The seven fields that matter

Certificates vary in layout, but the ones worth having share the same skeleton. Check each of these in order.

1. Compound name and batch (lot) number

The batch number is the whole point. It is the string that ties the paper to the carton in your hand. Cross-check it against the label — and against the batch lookup on our testing page, where every live batch is listed with its measured figure. A certificate with no batch number, or a batch number that doesn't match, tells you about some material. Not yours. How a batch number ties the pen in your hand to the sample that was tested — and what it cannot prove on its own.

2. Date of analysis

When the test was run. Not the manufacture date, not the date the PDF was generated, and not the day the web page was last edited. A date more than a year or so old on a compound with a stated shelf life shorter than that is a question worth asking.

3. Method

You are looking for the words HPLC (high-performance liquid chromatography) and, ideally, MS (mass spectrometry). HPLC separates the components of the sample and measures how much of the total signal comes from the main peak. MS confirms that the main peak has the mass of the molecule it's supposed to be. One without the other is weaker than both together — we've written a separate piece on what those two methods do and don't tell you.

4. Purity — and what it is a percentage of

Purity is typically reported as a percentage of the total chromatogram area that belongs to the target peak. So "99.3%" means: of everything the detector saw, 99.3% was the compound and 0.7% was something else — usually closely related impurities such as truncated or deamidated sequences. On a blend of two or more compounds there is no single purity figure to report, which is why a blend report has no purity percentage.

What it does not say is how much material is in the vial. A vial could be 99% pure and contain a third of the labelled quantity. Which brings us to —

5. Quantity, content or "assay"

Some certificates include a mass or content figure — the measured amount of peptide per unit against the label claim. Many don't, because it requires a reference standard and a calibrated method. If it isn't there, the certificate is silent on quantity and you should know that it's silent. This is where underdosing hides: 99% pure is not 10 mg, because identity, purity and quantity are three different measurements. Why that figure matters is clearest in the published retatrutide data, where content was the most common problem: is retatrutide fake? What independent testing has actually found.

6. The observed mass

On an MS-backed certificate you should see a measured molecular mass (often written as m/z or "observed mass") alongside the theoretical mass of the compound. The two should agree to within a small tolerance. This is the identity check: it's what stops a vial of something 99% pure from being passed off as a vial of the thing on the label.

7. Who did the work

A named laboratory, or the manufacturer's own quality-control department, with a way to identify the document (a report number, a signatory). "Third-party tested" without a name is a claim, not a fact. Manufacturer QC is not third-party — it is a real and useful document, but it should be described as what it is, which is why our testing page says so in plain terms.

When the named lab is Janoshik, three follow-up reads cover most questions: what a Janoshik report proves and what it doesn't, how to check a Janoshik report is real, and what it means when Janoshik verification is not working. If you would rather send a sample yourself, here is how to get a peptide independently tested in the UK.

The decoration

Things that look reassuring and mean little:

• A logo and a stamp. Both take thirty seconds to add.
• "Pharmaceutical grade." Not a defined term outside an actual pharmacopoeia monograph. What research-grade and pharmaceutical-grade actually mean.
• A purity of exactly 99.99%. Real chromatography of a synthetic peptide rarely produces that number; be more suspicious of it, not less.
• A chromatogram with no axis labels, no peak table and no retention times. A picture of a peak is not a result.

Six mismatches that should stop an order

1. Batch number on the certificate ≠ batch number on the carton.
2. Theoretical mass on the certificate ≠ the known mass of the compound. (Ten seconds on a chemistry database will give you the reference figure.)
3. A method line that says only "HPLC" when the page selling the product says "mass spec verified".
4. A certificate dated after the product page went live with that purity figure.
5. Identical certificates — same numbers to the decimal — across different batches or different compounds.
6. A vendor who will only send the certificate after payment.

If the certificate is a Janoshik report, the free Janoshik report checker reads the link or key in your browser and takes you to the report on the lab's own site, which is the version that counts.

What we send

Name a compound, and we'll forward the manufacturer's certificate for the batch you would receive — before you order, no account needed. The number on the product page is the number on that document. The testing page walks through what's on it, and is equally clear about which batches carry an independent Janoshik report — verifiable by key on Janoshik's site — and which are still waiting for one. Everything we supply is for laboratory research use only.