How to Read a Certificate of Analysis for Research Peptides
Every section of a research-peptide COA explained — identifiers, the seven result categories, what each number means, and what a certificate cannot tell you.
Published 2026-08-18. By Elevate Research Labs. · 4 min read
A certificate of analysis is the one document in this category that can be checked rather than believed — but only if you know what you're looking at. This walkthrough uses the structure of our own certificates, and every example below comes from a real, signed, currently-published one. You can pull any of them up as you read: every published lot is listed here.
First: what a COA is, and is not
A COA is a testing laboratory's signed record of what one specific lot was tested for and what the results were. Three consequences follow.
- It describes a lot, not a product line. A certificate with no lot number on it describes nothing, because there is no way to tie it to the vial in your hand.
- It is only as independent as its author. A certificate generated by the seller is a claim; a certificate signed by an outside laboratory is a record someone else stands behind.
- It reports what was measured on the analysis date. It is not a warranty about storage, handling, or anything that happened to the material afterward.
The header: identifiers
The top of the certificate carries the identifiers that make everything else meaningful.
- Lot number — ties the certificate to one production run. On our labels this looks like
BC10-2026-001. The lot on the certificate must match the lot printed on the vial; that match is the entire chain of custody from your bench back to the lab's record. - COA number — the certificate's own document ID, assigned by the laboratory.
- Accession number — the ID the laboratory gave the physical sample when it arrived. This is the key into the lab's internal records.
- Dates — received, tested, released. A certificate dated before the product line existed is a red flag worth acting on.
- Access code — the short code that resolves on the issuing laboratory's website to their copy of this record. This is the piece most certificates in the category are missing, and it is the difference between paperwork and proof. How to use it is its own article: verifying a COA with the issuing laboratory.
The seven result categories
Our release panel runs seven checks per lot. Here is what each one is, in the order it appears.
1. Purity (RP-HPLC)
The share of everything the chromatography detector saw that was the target compound, as an area percentage. When a certificate says a lot of BPC-157 assayed at 99.72%, it means all other detectable peaks together made up 0.28%. Purity answers "how much of this is one substance" — it deliberately does not answer "which substance," which is the next test's job.
2. Identity
Confirmation that the material is the named compound, established by comparison against a certified reference standard. A high purity number without identity confirmation is an incomplete story: a vial can be 99% pure something-else. Read purity and identity together, always.
3. Net peptide content
The measured mass of actual peptide in the vial, as distinct from the counter-ions, residual water, and salts that also carry mass in a lyophilized cake. A vial labeled 10 mg can measure above or below 10 mg of peptide proper — our certificates report the measured figure (for example, 10.46 mg against a 10 mg label) rather than grading it, because it is a disclosure, not a pass/fail test.
4. Fentanyl screen
A targeted screen confirming the absence of fentanyl. It exists because contamination cases in gray-market chemical supply made it worth ruling out explicitly, on the record.
5. Elemental impurities (ICP-MS)
Heavy metals — lead, arsenic, cadmium, mercury — measured by inductively coupled plasma mass spectrometry against specification limits. This is a spec'd test: the certificate shows the limit and the measured value side by side.
6. Sterility (by PCR)
A nucleic-acid-based screen for microbial contamination in the sampled material.
7. Endotoxin — and why it is reported, never "passed"
Bacterial endotoxins are measured per the USP's bacterial-endotoxins chapter, and this is the one line on the certificate that carries a number but no pass/fail grade. That is honesty, not evasion: acceptable endotoxin limits depend on the application and matrix, and no universal threshold applies to research-use material. A certificate that stamps PASS on an endotoxin value is claiming a judgment the laboratory has no basis to make. Ours reports the measured value and leaves the judgment where it belongs — with the researcher who knows the application.
What to do with all of this
Reading a certificate takes about ninety seconds once you know the layout:
- Match the lot number on the certificate to the lot on the vial.
- Check purity and identity together.
- Glance at the spec'd tests (metals) for limit-versus-result.
- Note that endotoxin is reported, not graded — and be suspicious of any certificate that grades it.
- Then stop reading and start checking: take the access code to the issuing laboratory's site and confirm the lab's own record matches the paper. Here is exactly how.
Every claim in this article is inspectable against our published certificates — start from any lot number. For laboratory research use only — not for human or veterinary use.
Grounded in Elevate's own published certificates and pages — every referenced lot and result is verifiable on the lot-verification page.
All materials are for in-vitro laboratory research use only — not for human or veterinary use.
Related reading
A Working Glossary of Research-Material Quality Terms
COA, HPLC, lot, accession, endotoxin, net peptide content — the vocabulary of research-material quality control, defined precisely and without marketing gloss.
2026-08-18 · 3 min read Verification & COAsHow to Verify Any Peptide COA With the Issuing Laboratory
A certificate is a claim until the laboratory that signed it confirms it. The 60-second verification workflow, the red flags, and a worked example you can run right now.
2026-08-18 · 3 min read