How to check a Chinese manufacturer's peptide COA before you rely on it

China COA verification: what you are checking

A certificate of analysis (COA) from a manufacturer in China records what the factory’s QC lab measured on a sample the factory chose. Verifying it means answering three questions:

1. Is it real? Did the stated lab issue this exact document?
2. Is it about this batch? Do the lot and the dates match what you bought?
3. Is it enough? Do the results, and the evidence behind them, answer the questions you need answered?

A COA can pass all three and still leave a gap: it describes the factory’s sample, not the vials that reach you. That is why many buyers check the COA first and then test their own vials.

What a factory COA usually contains

Factory COAs vary in layout. Most are one or two pages with a results table. The fields below are the ones you will see most often.

Field	What it tells you	What to check
Product name, sequence or formula	Which peptide the factory says this is	The formula and molecular weight match published values
Lot or batch number	Which batch was tested	It matches the lot on your vials and invoice
Manufacture and test dates	When the batch was made and tested	Test date after manufacture; no future dates
Appearance	What the powder looked like	White or off-white lyophilised powder is typical; note any colour
Identity by MS	The mass the factory measured	It matches the expected mass (table below)
HPLC purity	Main peak as % of UV peak area	The wavelength is stated and a chromatogram is attached
Peptide content	How much of the powder is peptide	Whether it is net peptide or total powder
Water, acetate or TFA	Other parts of the powder	These explain why the peptide weighs less than the powder
Conclusion and signature	“Conforms” and a QC sign-off	A named person and a stated specification

What a factory COA often leaves out

The fields that matter most for trust are the ones most often missing:

• Raw data. A full chromatogram with a peak table, and a mass spectrum with the observed mass. A number with no data behind it cannot be checked.
• The method. Column, wavelength and how peaks were integrated. ICH Q6A notes that identity by retention time alone is not specific; a mass result is stronger.
• The reference standard. Content in mg means little unless it was measured against a standard of known content.
• The basis. Whether content is net peptide or total powder. The EMA peptide guideline expresses assay on a counter-ion-free, anhydrous basis.
• Who tested it. The lab’s name and address. The WHO model certificate asks for the issuing lab, the specification and an authorised signature.
• Independence. A factory testing its own product has an interest in the result. That does not make the COA wrong, but it means the COA is not third-party evidence.

Six checks you can run yourself

1. Match the lot. The lot on the COA, the vial label and the invoice should be the same. A COA reused across batches is a known problem.
2. Check the dates. Manufacture, then test, then issue. A test date before manufacture, or in the future, means the document needs explaining.
3. Check the mass. Compare the reported molecular weight or observed mass with the expected value below. Small decimal differences are normal. A high-resolution result may give the monoisotopic mass instead, which for the GLP-1 peptides is about 2 to 3 Da below the average mass. A difference of tens of daltons points to a different or modified molecule.
4. Ask for the raw data. Request the full chromatogram, the peak table and the mass spectrum for this lot. A manufacturer that tested the batch has these files.
5. Check the arithmetic. The main peak’s area as a share of all peak areas in the table should match the purity printed. Content as % of label should match the mg and the label claim.
6. Verify any outside report at the source. If the COA names an outside lab, type the lab’s web address yourself and look the report up by its ID and key. Do not rely only on a link or QR code inside the PDF.

Peptide	Average mass, calculated from the published formula (Da)
Semaglutide	4113.58
Tirzepatide	4813.45
Retatrutide	4731.35
BPC-157	1419.54
Thymosin beta-4 (43 amino acids, N-acetyl)	4963.44
GHK (free peptide, without copper)	340.38

An outside LC-MS identity test reports the observed mass beside the expected mass, with the error in parts per million. That turns check 3 into a measured result.

How often COAs do not check out

COA problems are measurable. In a sweep on 23 May 2026, Disclosed Labs looked up 2,252 COAs on 8 labs’ own portals. The labs confirmed 2,092, or 92.9%. The other 160 were not confirmed by the issuing lab. Not every unconfirmed COA is forged, but none of them can be relied on until it is.

PeptideBenchmark flags anomalies such as one report reused across compounds, identical files across vendors, dates out of order, future testing dates and one lot shared across vendors. Peptigrity removes COAs whose figures do not match the lab’s own record. These checks cover lab reports posted by sellers. A factory’s in-house COA has no outside portal at all, which is why the raw data matters more.

For the warning signs in detail, see fake peptide COA signs and reused COAs. For the general method that works on any lab’s report, see how to verify a peptide COA.

When a supplier’s COA is not enough: test the batch

A COA is a good start. Test the batch when:

• It is the first order from a new supplier. You have no history to compare against.
• You will sell it under your own brand. Your name goes on the result.
• The COA has no raw data. Numbers alone cannot be checked.
• The results look too clean. Purity printed to three decimals with no chromatogram, or every batch at the same figure.
• The lot does not match. Or the parcel arrived late, warm or damaged.
• The order is large. The cost of a bad batch grows with the order.

When the outside result arrives, compare it with the COA line by line. Differences in purity of a point or two, or in content basis, are common and explained in factory COA vs an outside lab. A wrong mass or a large content shortfall is not.

How HK Analytical reports help with this

Our lab is in Hong Kong. Samples from mainland China reach us by cross-border courier. Our reports print the observed and expected mass, the full chromatogram and the content per vial against a reference standard, and every number comes with its raw data. Each report is a signed record with a private link that you can check on hkanalytical.com/verify, and it states how the client is related to HK Analytical. See the example report, read how to read a COA, or request a quote.