Quick answer
A COA is a lab's test report for one lot. Start by matching the lot number to your material, then read each result next to the method behind it. Last, check the lab name, test date and certificate ID.
A certificate of analysis (COA) is the record that describes a research material. If you're new to COAs, your first one can look like a page of codes and numbers. It's easier to read once you know where to look.
A COA isn't marketing, and it isn't a label claim. It's a lab's test report for one specific lot. Below, we walk through a COA field by field. Our example is a real published record from our Verification Center: BPC-157, lot PL-BPC10-0826-01, tested by Freedom Diagnostics.
What a COA is, and what it isn't
A COA is a lab's organized report on one batch of material. It records what was tested, how it was tested and what the instruments reported. It describes a test and its results, not what a compound does.
Quality guidance from ICH and the World Health Organization describes the same core contents for a COA. The list covers the material and its batch number, and each test with its acceptance limits and numerical results. It also covers the lab that did the work, plus a date and an authorized signature.
A COA is:
- A record tied to one lot or batch, not a general statement about the compound
- A summary of lab results from named test methods
- A document you can trace back to its origin: an issuing lab, a report ID and dates
A COA is not:
- A description of how a material behaves, or any claim about outcomes or use
- A permanent fact about the compound in general (a new batch gets a new COA)
- A substitute for checking the record against its original source
Keep that difference in mind as you read on. Every field below describes the document and the measurement, and nothing more.
Field by field
Six fields do most of the work. For each one, here's what it tells you and what to check.
- Lot number
- The batch the record describes. Check: it matches the lot on your material exactly.
- Product name
- The compound the lab says it tested. Check: it matches the product you're comparing it with.
- Test method
- How each result was produced, such as HPLC or mass spectrometry. Check: every result has a method next to it.
- Identity
- Whether the tested material is consistent with the named compound. Check: where the record reports it, it appears as its own line.
- Purity / result
- Of everything the method detected, the share that is the main compound, for that lot. Check: read it with its method and the identity line, not on its own.
- Lab and date
- Who did the testing, and when. Check: a named lab, a test date and a certificate ID you can look up.
The header: compound, lot number, certificate ID and dates
You'll usually find the header at the top of the page. It tells you which record this is and what it describes.
In our example, the header names the compound as BPC-157 and the lot as PL-BPC10-0826-01. That lot number anchors everything below it. The results apply only to that batch.
A complete record also carries a certificate ID and one or more dates. On the published BPC-157 record, the certificate ID is POLA2608310062 and the test date is September 2, 2026. The record also lists a net content of 11.91 mg, the laboratory's stated quantity for the tested sample.
Think of the certificate ID as the unique label for that one document. It helps when you want to point to the exact record or find it again later. A missing lot number or certificate ID is worth noting, because it makes a record harder to trace back to its source.
The results table, row by row
Most COAs are built around a results table. Each row is a separate test, and the columns tell you how to read it. A typical layout looks like this:
| Column | What it tells you |
|---|---|
| Test | What was examined (for example identity, main-component result or appearance) |
| Method | The test technique used for that row |
| Result | What the instrument or analyst reported |
| Specification | The expected range or limit, when one is listed |
Reading across a row (test, method, result) tells you more than the result alone. The method shows how the lab got that result.
Different rows may use different techniques. A result from one technique and an identity check from another answer different questions, which is why the method column matters.
The identity line and the result line
Two rows usually carry the most weight: identity and the main test result. They report different things, so they shouldn't be merged into one number.
The identity line answers a simple question: is this material what the label says it is? Where a COA lists an identity result, it reports whether the tested material is consistent with the named compound. That check is usually backed by a method suited to confirming molecular identity.
One detail is easy to miss: the on-page summary of the published BPC-157 record shows a test result but no separate identity line. Treat identity as present only where a record explicitly reports it.
The result line answers "of the components the method detects, what share is the intended compound?" On the BPC-157 record, the published test result for lot PL-BPC10-0826-01 is 99.62%. That figure comes from the lab's method and applies to that batch only.
Read it as a measurement of that lot, not as a permanent grade for the compound. A high figure means little unless the identity line first shows the right compound was measured. That's why the two lines are read together.
The lab, signatures and accreditation notes
A COA is only as traceable as the lab behind it. The footer or margin usually names the issuing lab, which here is Freedom Diagnostics. It may also carry an analyst or approver signature, a job title, and any accreditation or scope notes the lab chooses to include.
In this part of the page, look for:
- A clearly named lab, not just a logo
- A signature or approval line showing that a person or role stands behind the report
- Any accreditation or method-scope note, read for what it actually says rather than what you might assume
These details help you tie the document to a source that is accountable for it. If you see an accreditation note, read its exact wording and don't assume it covers more than it states.
What a COA leaves out
A COA answers a few specific questions. Here's what it can't tell you.
- Anything beyond the sample. A report describes only the sample that was tested. Labs accredited to ISO/IEC 17025 must state that results relate only to the items tested. When a supplier sends the sample, the results describe it as the lab received it.
- Anything about other lots. Each new batch needs its own record.
- What its methods can't see. Every method has a detection limit and blind spots. We cover these in What HPLC and mass spec mean on a COA.
- Approval or safety. A COA is not FDA approval or a safety assessment.
Checking the record against its source
The last step is making sure the COA is genuine and current, and not a reposted or altered copy. You do that by matching the record you have against the source that issued or published it.
A simple order to follow:
- Match the lot number on the COA to the lot on the material and the product record. They should be the same.
- Compare the certificate ID with the published version, where one is available.
- Check that the issuing lab named on the document is the source you expect.
- Find the same record in the source's own index, rather than trusting a loose file.
PolarLabs publishes its records in the Verification Center, where you can match a lot to its document. Going back to the source is what turns a COA from a piece of paper into a checkable record. When a product has no published record yet, its status reads Documentation pending.
Sources
- ICH guidelineICH Q7, Good Manufacturing Practice Guide for Active Pharmaceutical Ingredients — section 11.4, Certificates of Analysis
- WHO guidanceModel certificate of analysis
- NISTISO/IEC 17025 crosswalk: reporting of results (clause 7.8)
- FDAICH Q2(R2) Validation of Analytical Procedures
These sources explain what a certificate of analysis is, in general. They don't describe any PolarLabs product, and citing drug-manufacturing guidance here doesn't mean our products are made under it.