Certificates of analysis: what they prove and what they don’t

Home / Source Safety Education / Certificates of analysis

Source safety

Certificates of analysis: what they prove and what they don’t

Educational content only. Not medical advice. This article teaches how to read a certificate of analysis. It is not an endorsement of any seller, product, or pathway.
30-second summary
What it is A certificate of analysis (COA) is a lab document that reports what an analytical test actually measured in a given batch: identity, purity, potency, and sometimes sterility.
What it proves Only what it was asked to test for, on the specific sample it tested, on the specific day the test ran. Nothing more.
Why it matters A real COA from an independent lab is a green flag. An in-house COA, or no COA, is a red flag. Knowing the difference is half the source-safety skill.
What it cannot do A COA does not replace a prescription. It does not make an unregulated seller legitimate. It does not guarantee the vial in your hand matches the one that was tested.
My bottom line

A COA is necessary, not sufficient. Read it like you would read a receipt: trust the specific numbers, trust nothing else.

What a certificate of analysis actually is

A certificate of analysis is a lab-produced document that reports the results of analytical testing on a specific batch of material. It is standard in pharmaceutical manufacturing, in food science, in cosmetics, in almost every regulated material-handling industry. When a pharmacy or manufacturer releases a batch of anything, the batch is tested, and the results of those tests are summarized on a COA that references that batch by lot number.

In the peptide conversation, the term gets used loosely. A legitimate 503A pharmacy can produce a COA from its own lab (one type of document) or share a third-party COA from an independent analytical lab (a different, stronger type of document). A gray-market seller can generate a PDF with the word “COA” in the header that does not actually describe any testing at all.

Calling all three things “a COA” is how readers get into trouble. The document type is the same. The evidence behind the document is wildly different.

TakeawayA COA is a statement about what a lab measured. The value of the statement depends entirely on which lab measured it and how.

The four things a COA typically tests for

The four core categories are identity, purity, potency, and sterility. Each one answers a different question, and a complete COA reports on all four.

Identity

Is the substance in the vial actually the substance on the label? Mass spectrometry or similar analytical methods can confirm the peptide’s molecular weight and sequence. A pass on identity means the molecule in the sample is what was claimed. A fail means you were about to self-administer something else entirely.

Purity

What percentage of the material is the claimed compound, versus impurities, fragments, or degraded peptide. High-performance liquid chromatography (HPLC) is the standard method. Purity is usually reported as a percentage (for example, 98.5%). The missing percentage is everything else in the vial, and “everything else” matters when you are about to put it into your body.

Potency

How much active compound is actually present in the stated quantity. A vial labeled with a specific amount should contain that amount, measured by a validated assay. Potency testing confirms the label matches the contents.

Sterility

Is the material free from microbial contamination? For anything that will be mixed with sterile water and introduced into the body, sterility testing is not optional. USP <71> is the United States Pharmacopeia sterility test standard. A real pharmacy-produced COA references that standard (or an equivalent) explicitly.

Many gray-market COAs skip sterility entirely, because it is the most expensive test to run and the hardest to fake. That omission is diagnostic.

TakeawayA complete COA reports identity, purity, potency, and sterility. A COA that skips any of these is not complete. A COA that only reports one of them is barely a COA at all.

How to actually read a COA

A legitimate COA is boring. That is the first thing to internalize. It is a structured document with specific fields. Here is what to look for, field by field.

  1. Testing lab name and address. A real lab is a real company with a physical address, a website, and (in the US) a CLIA or equivalent accreditation where applicable.
  2. Sample identification. The compound name, the batch or lot number, and the date the sample was received by the lab.
  3. Test methods. HPLC for purity, mass spec for identity, USP <71> (or international equivalent) for sterility. Real methods are named.
  4. Specification versus result. Every test line should show the specification (the pass criterion) and the actual measured result. A line that only shows “pass” with no number is not a real test report.
  5. Date of testing. Not the date the COA was typed up. The date the test was performed.
  6. Analyst signature or initials. A real lab attributes the test to a specific analyst. A digital stamp with no name is a weaker signal.
  7. Lot and batch tracing. The lot number on the COA must match the lot number on the vial. If those numbers do not match, the COA is describing a different batch than the one you are looking at.

When readers send me a COA to review, I read those seven fields first. If three or more are missing or vague, the document is not doing the work a COA is supposed to do, regardless of what the top of the page says.

TakeawayLot number on the COA must match lot number on the vial. If they do not match, the COA is describing some other batch entirely.

Red flags in COAs

These are the specific tells that the document in front of you is not what it claims to be.

Green flags
  • Independent third-party lab with a verifiable address
  • Named analyst, signed and dated
  • Identity, purity, potency, sterility all reported
  • Specific numbers for each test line, not just “pass”
  • Batch/lot number matches the vial you have in hand
Red flags
  • “In-house testing only” (the seller tested themselves)
  • Missing or generic batch number
  • No analyst name or signature
  • Only one or two tests reported (often just purity)
  • “Pass” with no measured value
  • Lab name that does not appear anywhere on the public internet
  • COA visible on the product page but not downloadable with a batch-specific reference
  • COA dated months before the product was listed (recycled from an earlier batch)
Watch for this one“Independent lab” is sometimes a shell. If the lab’s address and the seller’s address resolve to the same business filing, the COA is not independent in any meaningful sense.

Why a COA alone is not enough

This is the part most readers skip, and it is the whole point.

Even a perfect COA from a top-tier independent lab tells you about the sample the lab tested. It does not tell you about the vial in your hand. Between the lab and your medicine cabinet, a number of things can happen: the lab tested Batch A and the seller shipped you Batch B; the vial you received was filled by a different facility than the one that was tested; the material was tested before it was repackaged; or the lab report is real but attached to a product line that has since changed formulation.

In a legitimate 503A workflow, a compounded preparation comes with documentation that ties the specific preparation to the specific patient it was filled for. The COA is one piece of a chain of custody. On the gray market, no such chain exists. There is the vial and there is a PDF, and the connection between the two is whatever the seller says it is.

A COA without chain of custody is a snapshot of a different day in a different room. Useful context, not a guarantee.

TakeawayA COA describes a sample. A prescription fulfillment describes your preparation. Those are different documents. A COA alone does not stand in for a prescription.

The difference between a COA and a full chain of custody

Chain of custody is the documentation that tracks a substance from raw material through final release to a specific patient. In pharmaceutical manufacturing and legitimate compounding, that chain includes:

  • Incoming raw material COA from the active pharmaceutical ingredient supplier
  • Internal batch records documenting how the preparation was compounded, by whom, and when
  • Release testing before the finished preparation leaves the pharmacy
  • A patient-specific fill record tying the finished preparation to a specific prescription
  • Storage and shipping documentation (temperature, time in transit)

A COA sits at one point on that chain. It is the most visible piece, which is why it is the piece gray-market operations imitate. The rest of the chain is what makes it real, and the rest of the chain is what gray-market operations do not have.

When I ask readers what they have been given by a seller, “a COA” is usually the only document. That is a signal. In a legitimate pathway, the COA is one document in a larger file. On the gray market, it is the only document, and it is doing a job it is not designed to do.


My honest take

Opinion, not evidence

This section is opinion. I am not endorsing the use of any peptide. Everything above this line is sourced from the United States Pharmacopeia, FDA compounding guidance, and analytical chemistry practice. Everything below is my personal perspective as one pseudonymous reader. Your situation is not my situation. Do not treat this as a recommendation.

The COA is the most over-weighted document in the peptide conversation. A reader will show me a COA and ask whether the product is safe, as if the document has the power to answer that question. It does not. The document tells you what one lab saw on one day. Whether the material in your actual vial matches that sample is a different question entirely, and that question is answered by chain of custody, not by the COA alone.

The COA is the most over-weighted document in the peptide conversation. It tells you what one lab saw on one day, not what is in the vial in your hand.

Once I understood that, I stopped treating COAs as proof of anything and started treating them as context. A COA from a known independent lab is a useful data point. A COA from a lab I have never heard of, or one that was clearly generated in-house, is less useful than no COA, because it is actively misleading.

A COA from a lab I have never heard of is worse than no COA at all. It invites false confidence.

For me, personally, the heuristic is this: I look at the COA as evidence of posture. A seller who produces a real one is showing me who they are. A seller who cannot, or who gives me something that falls apart under the seven-field read, is also showing me who they are.


Questions to ask any seller about their COA

These are the questions I use when readers forward me a sample COA and ask me to read it. Ask the seller these before you trust the document.

  1. Which independent lab performed the testing, and can I contact them to verify the report?
  2. Does the lot number on the COA match the lot number on the specific vial you are shipping me?
  3. Do the test results include identity, purity, potency, and sterility?
  4. What specification was used for the sterility test, and is the result reported as a specific value or only “pass”?
  5. What is the date the sample was actually tested, and what was the date it was collected from the batch?
  6. Can you provide the raw HPLC chromatogram or mass spectrum, not just the summary report?

What to do next

If you are learning

Read the 503A framework article

The regulatory pathway that gives a COA its meaning. Read this alongside the COA article.

Open the article →
If you are evaluating

Red flags in 60 seconds

The ten warning signs that tell you a seller is not a pharmacy, regardless of what their COA looks like.

Open the guide →
If you are preparing

Prepare for the clinician conversation

How to bring a COA, a compound name, and a short set of questions into a real clinical conversation.

Open the guide →

Sources

  • United States Pharmacopeia. General Chapter <71> Sterility Tests.
  • United States Pharmacopeia. General Chapter <621> Chromatography, including HPLC methods.
  • FDA. Current Good Manufacturing Practice (cGMP) resources for pharmaceutical preparation.
  • FDA. Compounding Quality Act and 503A compliance guidance.
  • International Council for Harmonisation (ICH) Q6A: Specifications for New Drug Substances and New Drug Products.

Regulatory and analytical references only. No seller names appear anywhere on this site.


Related reading

Source safety pillar

The 503A framework

The regulatory pathway behind a legitimate COA. The document makes sense once you understand the system that produced it.

Source safety read next

Red flags in 60 seconds

The fastest way to disqualify a seller before you read a line of their COA.

Healing peptide no human trials

BPC-157 monograph

A Category 2 peptide where the source-safety framework disqualifies every seller by construction. A useful case study.

The Peptide File provides educational content based on published research and publicly available regulatory resources. This article is not medical advice. The Peptide File does not sell, distribute, or facilitate the acquisition of any peptide compound. Always work with a qualified healthcare provider.
Scroll to Top