A specification sets the target
A specification defines the material and acceptance requirements, including relevant methods and reporting conventions. It may describe identity, marker content, physical characteristics and selected contaminant controls. It is a decision framework, not evidence that a particular lot was sampled or tested. An unsigned list of typical values should not silently become an approved purchase specification.
A COA reports a specific lot
A certificate of analysis should identify the sample or batch and the actual reported results with appropriate methods and units. Check that it matches the ordered material and agreed specification revision. A result expressed only as complies may require clarification where a numerical value is necessary for your acceptance decision or destination requirements.
Review gaps rather than inventing data
Ask about sampling, test frequency, subcontracted work and how untested attributes are controlled. A COA does not prove that every possible hazard has been excluded, and a management-system certificate is not a replacement for lot evidence. The existing document checklist helps request real records; this site does not generate fictional COAs or claim access to certificates not supplied by the business.
Approval trail
Retain the agreed specification revision and document who reviewed any deviation from the acceptance criteria.
Start with a document-to-lot match
Use these as review questions, not a claim that a supplied document already contains the answers: Does the sample identifier connect to the delivered batch and exact SKU, including plant part, extract form and carrier? Which specification revision was agreed at purchase, and which revision does the COA cite? Who sampled the material, when was it tested, and which laboratory issued or subcontracted each result? Request the traceability record rather than assuming that a generic example COA covers the shipment.
Read the method and reporting basis together
For every numerical line, ask for the method identifier and revision, units, actual result and matching acceptance criterion. Is the assay reported on an as-received basis or a dry basis? If a correction was used, request its calculation and the measured drying loss or water value used in it. Do not convert between bases from the label alone. A line called moisture requires clarification: WHO’s 1998 manual says LOD includes water and volatile matter.[1] If the specification calls for water by KF but the report gives LOD, flag a method mismatch; do not treat the percentages as equivalent.
Resolve a pass statement or method change
Where a numerical decision is needed, ask what lies behind “complies”, “not detected” or a less-than result. What detection or quantification limit applies, and is it adequate for the agreed criterion? For identity tests, ask which identification procedure and reference were used rather than demanding an irrelevant number. For a method change, request the reason, effective revision and evidence that results remain comparable for this matrix. Keep the discrepancy open until the responsible reviewer records an acceptance decision or requests suitable testing.
Acceptance is not a wider guarantee
Ask which attributes were tested on this lot and which rely on periodic testing or another control plan; record what remains unverified. Passing the agreed tests is not a market authorization, proof of every hazard’s absence or a shelf-life guarantee. Do not silently rewrite the specification to fit the result. Retain the authentic COA, agreed specification and documented disposition of deviations together; these review prompts are not invented batch results or a compliance declaration.