What a Certificate of Analysis Actually Proves
How to Read a Peptide Certificate of Analysis
Lab letterhead. A clean chromatogram. A number near the bottom: 99.2%. It looks like an answer — to a much narrower question than most readers assume.
Someone sends you a document. Lab letterhead. A graph with a tall clean peak. A second graph covered in spikes. A number near the bottom: 99.2%.
It looks like an answer. It is an answer — to one narrow question, asked about one batch of material, on one day.
The question it answers is does this vial contain what the label says? That's worth knowing, and most of what circulates in this market can't establish it. But that question is routinely confused with a much larger one — should anyone be taking this? — and a certificate of analysis has nothing to say about the second. The document is often deployed as though it did.
This piece is about reading one properly. It's the companion to Article 1, which established that a name doesn't identify a molecule. This is how you check what actually got named.
First: is this document about your material?
Before reading anything technical, check the boring fields.
A CoA covers a lot — a single production batch, identified by a lot or batch number. That number should appear on the document and on the vial, and they should match. A certificate displayed on a website with no lot reference isn't evidence about anything you might be holding. It's a sample of what their paperwork looks like.
It should carry a date. Testing is a snapshot. A purity figure from eighteen months ago describes the material as it was eighteen months ago, which for peptides — sensitive to heat, moisture and light — is a meaningfully different claim than describing it now.
It should name the laboratory. Not "third-party verified." The actual name of the facility that ran the analysis, ideally with a report reference number.
The four technical fields worth reading
1. Identity — usually by mass spectrometry
What the test does: measures the molecular weight of what's in the sample and compares it to the weight calculated from the claimed structure.
What it establishes: the main component has the expected mass. Given that peptides are built from a fixed set of building blocks, that's a real constraint. If the observed mass is wrong, something is wrong.
What it doesn't establish: mass does not uniquely determine a molecule. Two peptides built from the same amino acids in a different order have identical molecular weights. So do versions built with mirror-image amino acids. Mass spectrometry confirms the material is consistent with the claimed identity — it doesn't prove the sequence.
That distinction is usually academic and occasionally decisive. Sequence confirmation is a separate, more expensive analysis, and it's rarely present on a routine CoA. If a compound's whole identity claim rests on a specific sequence — and after Article 2, you'll know how much two residues can matter — mass alone is a weaker confirmation than it appears.
2. Purity — usually by HPLC
What the test does: separates the sample into its components, passes them by a detector, and reports the target peak as a percentage of everything detected.
Here is the part almost nobody reads correctly. That percentage is a share of what the detector saw, under particular conditions.
The detector works by shining light through whatever emerges from the column and measuring how much is absorbed. Different chemical structures absorb different colours of light, so the wavelength gets chosen to suit what's being looked for — and for peptides, the standard choice is one absorbed by the peptide bond itself. Wavelengths are given in nanometres, abbreviated nm, which is why certificates carry a notation like "at 214nm."
The consequence: things without peptide bonds — salt, residual water, some solvents — barely register. The detector is tuned to see peptide, so peptide is largely what it reports.
So "99% pure by HPLC" does not mean the vial is 99% peptide by weight. It means that of the peptide-related material detected, 99% was the target. The remaining mass of the vial isn't addressed by that number at all.
This isn't a trick, and it isn't unique to unapproved compounds — it's the standard convention, used the same way in legitimate pharmaceutical analysis. It's just consistently misread by people who assume a purity percentage refers to the contents of the container.
3. Salt form and net peptide content
Purified peptides are almost never isolated as the bare peptide. They come out of the purification process paired with a counterion — a small charged partner molecule that balances the peptide's own electrical charge and travels with it. Commonly that's trifluoroacetate, a leftover of the purification chemistry itself, or acetate, often used when the material is intended for biological work. A peptide paired with a counterion is described as a salt form, which is where that term on the certificate comes from.
The powder in the vial has then been lyophilised — freeze-dried, meaning frozen and placed under vacuum so the water leaves as vapour without ever melting, which is gentler on fragile molecules than heating. What remains is some combination of peptide, counterion, and residual water. The stated weight usually refers to all of it.
This is the item from Article 1's specification list that gets omitted most often, and it's the one that quietly breaks any calculation performed downstream. A stated weight without a salt form and a net peptide figure is not a quantity. It's an upper bound.
4. Appearance, and what else was tested
Physical description, solubility, water content. Individually unremarkable; collectively they indicate whether anyone treated this as a real analytical exercise or generated a document.
More informative is what isn't on the page. Which brings us to the important part.
What a certificate of analysis cannot tell you
Identity and purity testing sit in one category. Several other questions sit in categories that a standard CoA does not touch, and their absence is easy to miss precisely because the document looks complete.
Sterility. A separate test. Identity and purity analysis says nothing about whether the material carries viable organisms.
Endotoxin. Also separate, and less intuitive. Endotoxins are fragments of bacterial cell wall — they're not alive, so killing bacteria doesn't remove them, and they're notably heat-stable. Material can be sterile and still carry them. This test is standard for anything intended for injection and frequently absent from the documentation circulating around unapproved compounds.
Stability. Every figure on the page describes the material at the moment of testing. Nothing on a CoA speaks to what storage and shipping did afterwards.
Bioactivity. No line on any certificate of analysis states that the molecule does anything. Identity and function are different claims, established by entirely different work, and one of them is far harder. A compound can be exactly what it says it is and have no demonstrated effect in humans whatsoever — which, as Article 6 covers, describes a substantial part of this category.
And the whole clinical question. Whether the compound is approved. Whether there's a defensible dose. Whether it interacts with something else being taken. Whether any of it is appropriate for a particular person. None of that is analytical chemistry, and none of it appears on this document.
The point that matters more than the technical content
Documentation quality and clinical validity are independent of each other.
A flawless certificate of analysis — correct lot, named lab, method-specified purity, net peptide content stated — establishes that the vial matches its label.
It does not create evidence that the compound works. It does not establish a dose. It does not make an unapproved compound approved, or a discontinued one safe. It resolves the identity question, cleanly and usefully, and leaves every other question exactly where it was.
This gets conflated constantly, and the conflation does real work. A well-produced CoA carries an air of rigour that transfers, unearned, to claims the document never addressed. The reasoning runs: this is thoroughly documented, therefore this is sound. The first clause is about paperwork. The second is about pharmacology. They're connected only in the sense that failing the first makes the second moot.
So the right way to hold this: a good certificate of analysis is necessary and nowhere near sufficient. Its absence tells you to stop. Its presence tells you that you may now begin asking the harder questions — the ones this series is actually about.
Where this fits
This piece supports the first tool in the series — a name is not a molecule — by showing what it takes to establish which molecule you're dealing with.
It also sets up the last one. Article 9 takes a real protocol apart, and one of the things it finds missing is any identity information at all. You now know exactly what should have been there and what it would and wouldn't have proven.
← Back to Article 1: The Same Name, Two Different Drugs → Continue to Article 2: Two Amino Acids Separate Bonding From Blood Pressure
This article is educational and not medical advice. It does not recommend any compound, dose, protocol, or supplier, and nothing in it should be read as guidance on obtaining unapproved compounds. Its purpose is to help readers critically evaluate documentation they have been shown. Decisions about any therapy belong with a qualified clinician who knows your history.