A certificate of analysis is the most quoted document in the peptide world — and the least read. Most buyers glance at one number, the purity percentage, and stop. That number is real. It is also only one line of the report, and it answers exactly one question.

Here is what each line actually proves, using reports from my own catalog, because a standard you can’t see applied isn’t a standard.

Purity (HPLC). This is the headline number: what share of what’s in the vial is the compound on the label, versus related impurities. My Tesamorelin report from Freedom Diagnostics reads 99.70% purity by HPLC-UV. Strong number. It says nothing, by itself, about whether the compound is the right compound, how much of it is in the vial, or whether anything living or toxic came along for the ride.

Identity (mass spectrometry). A separate test. Purity tells you the main ingredient dominates; identity confirms the main ingredient is what the label claims. The same Tesamorelin report confirms identity by LC-MS. A vial can be 99% pure and 100% the wrong molecule — purity without identity is a number about something, not necessarily about your product.

Net content. Also separate. The Tesamorelin label says 10 mg; the lab measured 11.93 mg. My Retatrutide 20 mg report measured 21.61 mg. Overfill is common and usually harmless to know about — but you only know it if the report measures it. A purity percentage never tells you the milligrams.

Net content — and why I publish the uncomfortable number. My newest report makes the point better than I could. The KLOW blend that came back from Freedom Diagnostics this week reads 99.95% pure, identity confirmed for all four compounds — and the vial measured 74.41 mg against an 80 mg label. GHK-Cu came in at 45.21 mg instead of 50; BPC-157 at 8.39 instead of 10. I published those numbers anyway, on the product page and to every KLOW customer, because a purity headline would have hidden them. Net content is its own line. If a seller only shows you the percentage, you don’t know the milligrams — and neither do they.

Endotoxin. Not in the standard panel at all. Endotoxin — bacterial toxin that can ride along in a powder — is measured by its own test (the LAL assay), with its own report. My Retatrutide endotoxin result came back under a separate accession number from its purity report: no endotoxin detected. If a COA doesn’t list an endotoxin line, it wasn’t tested for endotoxin. Silence is not a pass.

Sterility. The line almost no research-peptide COA carries. True sterility testing is a separate, slow test — a 14-day incubation — and it is not part of a normal identity/purity/content package. That includes mine: the KLOW panel I pay for covers purity, identity, and net content — no endotoxin line, no sterility line, so I don’t claim either.

So the honest way to read any COA, mine included: check that the lab is independent and named, that the report ties to the lot in your hand, and then read every line for what it is — and notice the lines that aren’t there. A COA is evidence, not a halo. The sellers worth trusting are the ones who show you the whole document and tell you what it doesn’t prove.

Read the trial before you buy the vial. And read the whole COA before you trust the headline number.

Research use only. Not medical advice.