You ordered a research peptide. It arrived with a PDF. The PDF has a table of numbers, a graph with peaks, a handful of acronyms, and a purity percentage in bold. Most people look at that percentage, decide it seems high enough, and close the file.
That purity number, on its own, tells you almost nothing. A COA has six or seven distinct sections, each answering a different question about what is in the vial. If you only read one of them, you are flying blind on the rest. Here is how to read the whole thing.
What a COA actually is (and what it is not)
A Certificate of Analysis is a lab document that reports what was found when a specific batch of product was tested. The key word is “specific batch.” A COA applies to one lot number, produced on one date, tested under one set of conditions. It is not a blanket statement about the product in general.
It is also not an approval. A COA does not mean the product is pharmaceutical grade, FDA approved, or safe for any particular use. For research-use-only peptides, a COA means exactly one thing: a lab ran a set of tests on this batch and here are the results. That is it.
For serious research, COAs are part of basic quality control and record-keeping. They let you trace a result back to a specific batch of material and verify that the material met your specifications before you used it. Treat them as documentation, not decoration.
The first thing to check: the header
Before you look at a single test result, check the header. This is where you confirm that the document in your hand corresponds to the vial on your bench.
Look for the product name and, if it is a peptide, the amino acid sequence. A COA for BPC-157 should list the sequence (Gly-Glu-Pro-Pro-Pro-Gly-Lys-Pro-Ala-Asp-Asp-Ala-Gly-Leu-Val) so you can confirm you are looking at the right molecule, not just the right label.
Next, match the lot number on the COA to the lot number on your vial. If they do not match, the rest of the document is irrelevant. You are reading test results for a different batch.
Finally, note the lab name, address, and issue date. A real COA identifies who performed the testing and when. If that information is missing, or if the lab does not appear to exist when you search for it, that tells you something important about the vendor.
Identity: is this the right molecule?
The identity section answers the most basic question: is this actually the peptide it claims to be? A vial labeled BPC-157 could contain a completely different sequence, and without an identity test, you would have no way to know.
The standard tool for this is mass spectrometry. Think of it as a very precise scale for molecules. The instrument measures the molecular weight of whatever is in the sample and compares it to the expected weight for the target sequence. On most COAs, you will see this as a line or table: “Calculated: 1419.5 Da; Found: 1419.6 Da.” If those numbers match within a small tolerance, the molecule is confirmed.
Some higher-end COAs include additional identity checks like amino acid analysis or peptide sequencing, but mass spectrometry is the baseline. If a COA does not include an MS result, there is no proof that the contents are what the label says.
A simple way to think about the relationship between the two main tests: HPLC tells you how much of the vial is your target compound. MS tells you whether that target compound is actually the correct molecule. You need both.
What that purity percentage actually means
This is the number everyone jumps to: “Purity by HPLC: 98.7%.” It is important, but it means something specific and narrower than most people assume.
HPLC works by pushing a liquid sample through a packed column at high pressure. Different compounds travel through the column at different speeds, so they exit at different times. A detector at the end records a chromatogram: a graph with peaks, where each peak represents a different compound in the sample.
The tall peak is your peptide. The small peaks around it are impurities: truncated sequences, synthesis byproducts, degradation products. Purity is calculated as the area of the main peak divided by the total area of all peaks, expressed as a percentage. “98.7% purity” means 98.7% of the peak area belonged to the target peptide under that specific method.
That number is method-dependent
Look for the actual chromatogram on the COA, not just the number. The graph is much harder to fabricate than a typed percentage, and it gives you real information: how clean the separation was, whether there are significant impurity peaks, and whether the baseline looks normal. A purity number with no chromatogram behind it is a claim without evidence.
Purity is not the same as content
This is the section that catches the most people off guard. A peptide can be 99% pure by HPLC and still deliver less active material than you expect. The reason is that HPLC purity measures the peptide relative to other organic compounds. It does not account for everything else in the powder.
Peptide powders are almost always lyophilized, and that dry powder contains more than just peptide. There is residual water, residual salts from the purification process, and counter-ions like acetate or TFA that tag along to balance the peptide's charge. All of this adds mass to the powder without adding active material.
Some COAs report net peptide content (sometimes labeled “assay” or “content”), typically as a percentage by weight. In practice, this number often falls between 60% and 90%. A vial labeled 10mg with 80% net peptide content contains about 8mg of actual peptide. The rest is water, salt, and counter-ions.
The 10% math problem
Think of it this way: purity tells you “how clean.” Content tells you “how much active material you actually have to work with.” They answer different questions, and you want the answer to both.
The safety section most people skip
Below the purity and identity data, most COAs have a section that covers contaminants, residual chemicals, and physical properties. This is the part of the document that forum threads almost never mention. For in vitro research, it can matter more than the purity number.
Endotoxins are bacterial cell wall fragments that survive sterilization. They are measured by the LAL test and reported in endotoxin units per milligram. If you are running cell-based assays, endotoxin contamination can activate immune pathways and produce results that look biological but are actually artifacts. A COA that reports endotoxin levels, even as a simple “pass” against a threshold, is telling you the vendor thought about this. One that does not mention endotoxins at all is telling you something too.
Residual solvents are chemicals left over from synthesis and purification. Acetonitrile and DMF are the usual suspects in peptide manufacturing. COAs that test for these will list them individually with measured levels and a specification limit, often following ICH guideline thresholds.
Moisture content, usually measured by Karl Fischer titration, tells you how much of the powder weight is water. This connects directly to the net peptide content discussion: more moisture means less active peptide per milligram of powder.
You will also sometimes see appearance (“white to off-white lyophilized powder”), storage conditions (typically -20 °C, protected from light), and a retest or expiry date. These are not glamorous data points, but they tell you whether the product was characterized properly and how long you can expect it to remain stable.
Quick red flags
We covered the deeper side of COA verification in our article on peptide testing. But even at a glance, certain things should make you pause.
No lot number, or a lot number that does not match your vial. No lab name or contact information. A purity percentage with no method details and no chromatogram. No mass spectrum for identity confirmation. Copy-paste errors or values that do not make physical sense (purity over 100%, negative moisture content). A single generic template reused across every product a vendor sells, with nothing batch-specific.
A trustworthy peptide COA shows, at minimum, HPLC purity with a chromatogram and MS identity confirmation, tied to a traceable batch number and issued by a named lab. Every COA we publish includes all of this, plus the raw data and a verification key you can check independently. If a vendor cannot provide at least the basics, the question is not whether you trust them. The question is what, exactly, they are asking you to trust.