How to read a cannabis COA
A cannabis certificate of analysis (COA) should help you reduce uncertainty, not create false confidence. Many teams skim a COA for one or two values, then assume the product is low risk. A stronger approach is to read the COA as a structured record: identify the lot, confirm sample relevance, review panel scope, evaluate measurement context, and capture any unanswered questions before procurement approval.
Step 1: Confirm identity and traceability
Start with basic alignment checks. If the tested lot, production date, or product category does not match what is being sold, every downstream value is less useful. COAs should map cleanly to the specific inventory under review.
- Lot/batch ID should match invoice and packaging records.
- Sample collection and report dates should fit production timelines.
- Laboratory name and report identifiers should be auditable.
Step 2: Check panel coverage, not just outcomes
Passing values only matter if the right analytes were tested. Different product forms carry different contaminant and process risks. A complete review asks whether panel scope is appropriate for the product format and market context, then documents any out-of-scope areas before approval.
Step 3: Read results with measurement context
Values near an action limit deserve more scrutiny than comfortably separated values, even if both are technically passing. Also watch for reporting quirks: non-detect notation without clear limits, unusual unit presentation, or missing method details on critical analytes. These issues do not always invalidate a COA, but they do affect confidence.
Step 4: Evaluate data consistency over time
A single COA is a snapshot. Procurement decisions improve when you compare multiple lots and look for stability. Repeated volatility, frequent corrections, or inconsistent panel availability can indicate process control problems even when no single report looks alarming in isolation.
Step 5: Capture unresolved issues in writing
Approval meetings move quickly. If unresolved questions are not logged, teams can accidentally treat assumptions as facts. Maintain a short COA review memo for each supplier that records exceptions, follow-up requests, and approval conditions. This improves accountability and makes future requalification easier.
Common COA review errors
- Approving based only on potency or one headline metric.
- Ignoring mismatch between tested matrix and marketed product.
- Treating “third-party tested” as a substitute for panel details.
- Failing to save the exact report version used for approval.
Practical review template
Use a fixed sequence for every COA: identity, scope, values, variance history, unresolved issues, and decision. Repeatability matters more than complexity. The objective is not perfect scientific interpretation; it is defensible procurement control with a clear evidence trail.