Testing guide
How to read a COA
A certificate of analysis (COA) is a lab's report on one sample. Start with what a complete report should include, then read it for four things: who tested it, exactly what was tested, which tests were run, and whether the results match the label.
What you should get when you pay a lab
Say you bought a vial sold as a GLP-1 and paid a known lab to test it. What comes back should let you check three things yourself: that the lab really tested your sample, which tests it ran, and that the results rest on real data. A complete report includes everything below.
| You should receive | Why it matters |
|---|---|
| A unique report number and a way to verify it: an online lookup, QR code or verification code on the lab's own site | Proves the lab issued this report and that nobody has edited it since. Without it, a PDF is just a document anyone could make. |
| Your sample identified: your order or sample ID, the description you gave, the date received, ideally a photo of your vial as it arrived | Ties the result to the vial you sent, not to a sample from someone else. |
| Each method named: for example LC-MS for identity, HPLC-UV for purity and quantity, LAL for endotoxin | Shows which questions were actually answered. "Purity 99%" alone doesn't say identity was tested. |
| Identity written out: the expected mass of the molecule, the mass found, and the mass spectrum | The only result that shows the molecule is the one on the label. See understanding lab results. |
| The HPLC chromatogram, the full run from start to end, with a peak table (time, area and area % of every peak) | Shows how the purity figure was worked out and that no peaks were cropped or left out. |
| How the quantity was measured: the reference standard used (source and lot) and the result per vial, in mg | Milligrams without a named standard are an estimate, not a measurement. |
| Your report number or sample ID printed on the chromatogram and spectrum | Stops a lab or seller from attaching data from a different sample. |
| Dates and a named person who approved the report | Someone at the lab takes responsibility for the result. |
| Accreditation details, if the lab claims any: the ISO/IEC 17025 certificate number and the accreditation body | Lets you check the claim and whether these methods are covered (see below). |
A PDF with only purity and milligrams: can you be sure?
No. A page that says "Purity 99.1%, 9.8 mg" is a claim, not evidence. On its own it doesn't tell you:
- Whether the molecule was identified. Purity measures how clean the main peak is, not which molecule it is. A vial of the wrong peptide can be 99% pure.
- Whether a reference standard was used. Without one, the milligrams may be worked out from purity or powder weight rather than measured.
- Whether a test was run at all. Without a verification code and the raw data, you can't tell a real report from an edited or invented one.
What to do: verify the report number with the lab, then ask for the identity result with expected and found mass, the mass spectrum, the full chromatogram with its peak table, and the reference standard used. A lab that ran the tests can send these quickly. If it can't, treat the result as unconfirmed.
Is the lab in good standing?
- Find the lab yourself. Look up its website and contact details independently; don't rely on links or emails in the PDF, which a forger can control.
- Check it is a real business. A registered company with a physical address, listed in its country's company register.
- Check any accreditation claim. ISO/IEC 17025 is the standard for testing labs. Look the lab up in the directory of the accreditation body it names; the ILAC signatory search lists the recognised bodies in each country. Check the scope too: accreditation covers specific methods, and may not include peptide identity or quantity.
- Don't rule out unaccredited labs. Many labs that test peptides for consumers have no accreditation. That doesn't make their results wrong, but it makes verification and raw data more important.
- Look at its track record. Blind tests from user communities, where the same vial goes to several labs, show whether a lab's results agree with others.
- Ask a question. A real lab can tell you which instrument and method it used, and send the raw data for your sample.
- ILAC MRA signatory search: accreditation bodies recognised to accredit testing labs to ISO/IEC 17025, by country
- ILAC: accredited conformity assessment bodies: accreditation certificates include the lab's scope of accreditation
The parts of a COA
12 Sample Street, City · [email protected]
- Lab name and contact. A real, independent lab with an address and a way to contact it. A COA with only a seller's logo is not a lab report.
- Report number and verification code. Use them on the lab's own website, or email the lab, to confirm the report is genuine and unchanged.
- Dates. When the sample arrived, was tested and the report was issued. An old report says nothing about a new batch.
- Sample and batch. What was submitted, as described by the person who sent it, and its batch number. It must match the batch on your vial.
- Methods. Which tests were run. LC-MS confirms identity; HPLC alone measures purity but is weaker evidence of identity.
- Results. Identity, purity, quantity, and sometimes endotoxin and sterility, each with a unit. See understanding lab results.
- Raw data. The chromatogram and mass spectrum behind the numbers. A COA with numbers but no data is harder to trust.
- Approval. A named person at the lab who signed the report off.
Who asked for the test matters
| Type of COA | What it shows | How much to trust it |
|---|---|---|
| Manufacturer's own COA | The maker's in-house test of its own batch | Not independent. Useful only if the maker is regulated and the batch matches. |
| Seller-commissioned third-party test | An independent lab tested a sample the seller chose and sent | Better, but the seller chose the sample. Verify the report with the lab. |
| Your own test, or a blind community test | An independent lab tested a vial bought like any customer would | The strongest evidence, for that vial. |
Red flags
- No lab name, address or contact, or a lab you can't find anywhere.
- No report number, or a number the lab's website can't verify.
- The batch number on the COA doesn't match your vial, or isn't given at all.
- The same COA shown for several batches or products.
- Purity only, with no identity test and no quantity.
- Results with no units, no chromatogram, or a chromatogram cropped to hide the edges.
- Fonts, spacing or alignment that change partway through the document, a sign of editing.
- A report more than a year or two old for a product sold today.
A COA covers one sample
Even a genuine, verified COA tells you about the sample the lab received. Your vial may come from a different batch, or have been stored badly since. If you need certainty about your own vial, see testing options.