A Certificate of Analysis is not the test itself. It is the release summary of work that may take 3 to 5 working days and several separate test runs. In quality operations at Leumas, I treat the CoA as the final record, not a substitute for the underlying testing.
Analytical testing in supplement manufacturing runs through 3 stages: incoming raw material checks before production, in-process monitoring during the run, and finished-goods testing before release. Each stage answers a different question, so passing one does not replace the others.
Before production: incoming materials
Each raw material entering production is tested or checked against supplier documentation before approval. The protocol depends on the material category and the risk associated with it.
For identified actives such as vitamin C, also called ascorbic acid, or magnesium glycinate, incoming checks generally include identity confirmation and a purity assay. Identity confirms the material is what it claims to be; the assay measures active content.
Botanical extracts need a broader check. Identity may be assessed through appearance, odor, and color, thin-layer chromatography, or microscopic comparison with a reference sample. These checks help distinguish the expected botanical from a material that only looks similar.
Excipients, including microcrystalline cellulose, magnesium stearate, and silicon dioxide, are checked for identity and compliance with the applicable food-grade or pharmaceutical-grade specification. Capsule shells are checked against their approved specification before use.
In-process testing during production
Once materials are approved, monitoring continues through the run. For capsule filling, the main in-process control is weight. Checks are taken at defined intervals, typically every 15 minutes, to confirm that fill weights remain within the approved tolerance.
A fill-weight result outside tolerance triggers a line hold. The team investigates possible causes, such as changed material flow, hopper vibration, or tooling wear, before restarting. Units made during the deviation are isolated and assessed rather than automatically released.
For powder blends intended for sachets or tubs, blend uniformity is checked before filling starts. Samples from multiple blender locations are tested for active distribution, helping confirm that the mixed batch is consistent.
Finished-goods release testing
The finished-goods panel determines whether a batch may be released for distribution. The panel depends on dosage form, active ingredients, and customer requirements. A standard panel can include potency, microbial limits, heavy metals, and disintegration or dissolution.
Potency or label claim assay. For vitamins, minerals, and botanical marker compounds, HPLC or UV-spectrophotometric testing measures active content per unit against the label claim. The approved specification may express the result as 100% to 110% of claim.
Microbial limits testing. Finished supplements are checked for total aerobic microbial count, total yeast and mold count, and the absence of specified pathogens. The applicable limits depend on the product and its specification.
Heavy metals. Finished supplements are tested for lead, arsenic, cadmium, and mercury against the applicable FSSAI and international requirements. Testing uses ICP-MS, meaning inductively coupled plasma mass spectrometry, or another specified method.
Disintegration or dissolution. Capsule disintegration testing checks that the shell opens and releases its contents within the time set in the formula specification. The test is performed using the applicable USP method.
What a CoA does and does not certify
A CoA does not verify future shelf-life performance. It documents that the batch was tested and met its specifications at release. It records the result at that point in time, not a guarantee that the product will stay within specification throughout its stated shelf life.
Stability testing places samples under controlled temperature and humidity. Long-term storage is typically 25 degrees Celsius at 60 percent relative humidity, while accelerated storage is 40 degrees Celsius at 75 percent relative humidity. Samples are assessed over the intended 12- or 24-month shelf-life period, with 7-day observations used where the study protocol calls for them.
Some retailers or export customers request a CoA from an accredited independent laboratory instead of the manufacturer's in-house lab. Third-party testing adds cost and can add 7 working days, but it may be required by the customer specification.