You already know the percentage is the line people quote first. And you already know it tells you the least until the lot matches. When you open a peptide certificate of analysis, start with the identifiers. The purity figure can wait.
A shorter checklist lives at how to read a certificate of analysis. This page is the expanded version. Use this URL when you want the red flags in one place.
Match the lot before anything else
Picture the label and the PDF in the same glance.
- The lot or batch number on the certificate has to match the lot printed on the vial or its label.
- The compound name on the report has to match the name on the label. A near spelling is not the same material.
- A sample identifier shows whether the file belongs to one shipment or has been reused across batches.
If the lot on the paperwork and the lot on the container differ, the document does not describe the container you have. Read that again before you go back to the percentage.
Laboratory name and report ID
Notice the laboratory name before the chromatogram gets your attention. A traceable report names the laboratory and carries a report identifier you can write down: a report number, a job number, or an equivalent ID the laboratory would recognize.
A logo with the header cropped off is not the same thing as a name you can look up. When the laboratory line is empty, you do not yet have a third-party report. You have a file that has not said who ran the test.
Verify the report with the laboratory
Once the report ID is in front of you, the next click is the laboratory's own site. Janoshik, for example, publishes a verification check so a report identifier can be compared with the laboratory's record rather than with a PDF that anyone can edit. Use the verification page that laboratory hosts. A seller's screenshot of a verification result is a claim about a page. The page itself is the check.
If the laboratory does not publish a lookup, you still have the name, the date, and the sample ID. Those are the lines you keep. A missing lookup does not invent a match.
Dates
You will read three dates when the file has them: the date the sample was received, the date the test was run, and the date the report was issued. A report date that sits years away from the lot you are holding is a question, not a comfort.
Mismatched dates are a red flag. So is a single old report stretched across a catalog of later batches. The date tells you which sample was in the room. It does not travel forward on its own.
HPLC and mass spectrometry answer different questions
HPLC, often reported as an area percent, is a chromatography result. It can support a purity figure for the peak the method measured. Mass spectrometry is the identity line: it speaks to whether the material matches the stated sequence or mass. One of those results does not replace the other.
When you see only a percentage, ask which method produced it. A purity figure with no method is a sentence until a report defines it. Identity without a method name is the same kind of sentence.
Endotoxin and sterility
Endotoxin and sterility, when they appear, are additional tests on the sample that was submitted. Their presence tells you those assays were reported. Their absence does not mean the untested property passed. Read the section header, the method, and the lot, then stop. The file does not become a broader claim because two extra rows exist.
Quantity and purity are not the same line
Purity is about the composition the method measured. Quantity, or the amount the label states, is a different claim. A high purity figure does not tell you the container holds the labeled amount, and a labeled amount does not tell you the purity. When a report includes both, read them as two rows. When it includes only one, notice which one is missing.
Red flags
Keep this list next to the file:
- No laboratory name.
- A report ID that shows up on a different lot, or a file reused across batches.
- An edited PDF: cropped headers, altered numbers, or a logo pasted onto a page that does not match the laboratory's layout.
- A generic certificate with no lot, no date, and no sample ID.
- Dates that do not match the container, or a report date that cannot belong to the batch in your hand.
One clean row does not validate a catalog. Verify the batch-specific certificate for the container you have.
For the independence question, continue to how to evaluate a supplier's third-party testing.