How to Read a Certificate of Analysis as a Brand Owner
Most brand owners file certificates of analysis without reading them, which defeats the purpose of having them. A COA is not paperwork. It is the evidence that your specification was met for a specific batch. Here is how to read one in five minutes.
What a COA Is For
This guide is for brand owners receiving certificates of analysis from herbal, Ayurvedic or supplement manufacturers.
A certificate of analysis states what was tested on a specific batch, by what method, and what the results were against stated limits. Its purpose is to let you verify that the batch you received meets the specification you agreed.
That verification only happens if someone compares the COA against the specification. Filing it unread provides a record that a document existed, which is not the same thing and will not help you if a problem arises.
Section by Section
What each part should contain and what to check.
| Section | Should Contain | Check |
|---|---|---|
| Product identification | Product name, your specification reference | That it matches your product, not a generic name |
| Batch number | The specific batch | That it matches the stock you received |
| Manufacturing date | Date of production | Consistency with expiry and your receipt date |
| Expiry or best before | Date | That it reflects your substantiated shelf life |
| Test parameters | Each test run | That every test in your agreed panel appears |
| Methods | Named method per test | That methods are named, not implied |
| Specification limits | The limit for each parameter | That limits match your agreed specification |
| Results | Actual values, not just pass | That values are given, and where they sit in range |
| Test date | When testing was performed | Plausible relative to manufacturing date |
| Authorisation | Signature or approval | That someone in quality has signed it |
Results stated as “complies” or “pass” without actual values are the single most common weakness. A value near the edge of specification is information you want; “pass” hides it.
What Should Prompt Questions
Six patterns worth querying.
- Missing tests from your agreed panel, which is the most common gap.
- Results given as pass or complies rather than actual values.
- No test methods named, which makes assay figures hard to interpret.
- Limits on the COA that differ from your agreed specification.
- Identical values across consecutive batches, which is implausible for natural materials.
- A test date before the manufacturing date, or implausibly soon after.
Trend Rather Than Single Batch
The most useful analysis is across batches, not within one.
Keep a Simple Log
Record key values by batch in a spreadsheet: assay, moisture, microbial counts. Patterns appear quickly.
Watch Drift
Values trending toward a specification limit over several batches indicate a process or sourcing change worth asking about.
Watch Uniformity
Natural materials vary. Identical figures batch after batch suggests reporting rather than testing.
Correlate With Complaints
When a complaint arrives, check the batch’s values. Patterns between complaints and particular batches are informative.
A brand owner with a batch log is in a much stronger position in any quality discussion than one with a folder of unread PDFs.
Practical Checklist
Do this for every batch.
- Confirm the product name and batch match what you received.
- Check every test in your agreed panel is present.
- Confirm methods are named for each test.
- Confirm limits match your specification.
- Record actual values, not pass statements.
- Check dates are plausible and consistent.
- Confirm quality authorisation is present.
- Log key values in your batch log.
- Query anything missing or inconsistent in writing.
- File the COA with the batch record in the product file.
Want Certificates You Can Actually Verify?
Share your product and our team will provide per-batch documentation with methods, limits and actual values stated.
Frequently Asked Questions
What is a certificate of analysis for?
To let you verify that a specific batch meets the specification you agreed. It states what was tested, by what method, and the results against stated limits. That verification only happens if the COA is compared against the specification rather than filed unread.
What is the most common weakness in a COA?
Results stated as “complies” or “pass” rather than actual values. A value sitting near the edge of specification is important information, and a pass statement hides it. Always request actual figures.
Why do test methods need naming?
Because different methods produce different numbers for the same material, particularly for marker assays in botanical extracts. An assay figure without a named method cannot be interpreted or compared across suppliers.
What should make me suspicious?
Missing tests from the agreed panel, results given as pass rather than values, no methods named, limits that differ from your specification, identical values across consecutive batches, and test dates that are implausible relative to manufacturing dates.
What is more useful than reading a single COA?
A batch log recording key values over time. Drift toward a specification limit indicates a process or sourcing change, implausible uniformity suggests reporting rather than testing, and correlating complaints with batch values is often revealing.