How to read a peptide Certificate of Analysis (COA)

A Certificate of Analysis is a document from a testing lab that says "we tested this batch of this product and here is what we found."

Vendors include a COA to signal that they take quality seriously. Most buyers glance at it, see a number that looks high, and move on. This post is for the buyer who wants to actually read one.

The five things that matter

1. The testing lab is named

The COA should name the lab at the top. The common ones in the peptide world are:

• Janoshik Analytical (Czech Republic) — the reference most trusted by the research-chem community
• Finnrick — often used by vendors who also supply finished-dosage product
• Cheminoz — less common, credible
• Vendor's own internal QC — this is not a COA. It is a vendor's word.

If the COA is "internal" or unnamed, treat it as marketing, not data.

2. The batch or lot number matches the product you received

Your product has a batch number somewhere (on the vial, on the box, on the paperwork). The COA references a batch number. Those numbers should match.

If the vendor sends you a generic COA that says "tirzepatide, typical purity 98.5%" without a batch match, that is a template, not a test result.

3. The compound identification

Two tests together:

• Mass spectrometry (MS) confirms the molecular weight of the compound. For tirzepatide, you want to see something close to 4813 Da. If it is off by more than a few, the molecule is wrong.
• HPLC (high-performance liquid chromatography) measures purity as a percentage of the peak area that is the compound you paid for versus contaminants.

A COA with only one of these is partial. A COA with neither is theatre.

4. The purity number

The industry convention: 98% or higher is good, 95 to 97 is acceptable with caveats, below 95 is suspect. The remainder is "impurities" which could be anything from degradation products to synthesis byproducts.

Two gotchas:

• A 99.8% purity on a non-peptide carrier (e.g., the COA tested mannitol instead of the active compound) tells you nothing useful. Check the compound name.
• Very high purity (99.5%+) on a cheap compound is a flag. Manufacturing cost scales with purity. If the price per vial is half the market average and the purity claim is higher than the market average, one of the two is wrong.

5. The date of analysis

A COA from two years ago on a compound made six months ago is not informative. The batch date on the product should be within a few months of the analysis date on the COA. If the analysis is dated before the batch date, the document is fabricated.

The three things that do not matter

Endotoxin test result below detection limit

Most COAs include an endotoxin test. The result is almost always "below detection limit" because the method is stringent. Unless the vendor is shipping clearly contaminated product (which the endotoxin test would catch but is rare enough to not be the bottleneck), this line is not the discriminator.

"Sterility" claim with no method

Sterility testing is a defined pharmaceutical procedure. A COA that says "sterile" without specifying USP <71> or equivalent is making a marketing claim, not a scientific one. The sterility of a research-chem product is a function of the vendor's handling, not the COA line item.

The color photograph of the powder

Some COAs include a photo of the product. The photo tells you nothing about identity or purity. It is branding.

The one question

Can you send me the COA for the specific batch number on the product you would ship me?

A serious vendor says yes within 24 hours, sometimes instantly via a scan on the product page. An operator playing COA theatre will send you a generic document that does not match your batch, or will stall, or will cite a reason why they cannot.

If the answer is anything other than "yes, here it is," walk away.

What we do with COAs at Panya

Vendors in our rubric get 0 to 3 on the COA signal:

• 0: No COA provided, or generic/unnamed-lab document only
• 1: COA provided but batch match is weak or analysis date is stale
• 2: Named lab, batch match, both MS and HPLC, purity ≥95%
• 3: Named lab, batch match, both MS and HPLC, purity ≥98%, dated within 90 days of batch production

A score of 2 or 3 is table stakes for the Featured tier. A score of 0 or 1 is disqualifying unless every other signal is exceptional.

If you are buying

Take our quiz and we email you a vendor match that has passed the COA check. Two minutes, no list.