Checklist · Cannabis · Laboratory Testing

Certificate of Analysis and Lab Testing Checklist: Illinois and Missouri Sampling, Release, Failures, and the 2026 Label Data

A certificate of analysis proves something about a tested sample. This checklist traces a batch from sampling through the result, the release decision, the label, the tracking system, and the dispensary, under Illinois’s testing rule and Missouri’s lot-based testing rule.

By Lincoln Square Law · 8-minute read

🖨 Print this checklist

What a COA does and does not prove

A certificate of analysis is evidence about a sample. It does not prove that the right batch was sampled, that the laboratory was authorized, that the product stayed unchanged after sampling, that the correct limits were applied, or that the right certificate is attached to the units being sold. A testing program has to control the whole path from batch definition to retail unit.

Start with a package ready for transfer or sale and trace it back to its batch, its tracking record, the sampling event, the laboratory report, and the production record. Then reverse the test: choose one report and find every package that relies on it. A mismatch in either direction is a compliance problem even if each document looks complete.

Batch, sample, and laboratory

• Define the batch or lot and lock its production record before the sample is taken.
• Confirm the laboratory is currently approved or certified in the state for the testing involved, and keep the evidence in the vendor file.
• In Illinois, have an approved laboratory employee select a random sample of 7 to 15 grams from different parts of the batch, and keep cured flower batches at or below 20 pounds.
• In Missouri, have the testing licensee sample at the harvest or process lot level with an employee of the licensee observing, using a random sample and a chain-of-custody record.
• Do not select the best material, dry or pre-test samples, or alter sample integrity in any way.
• Keep the sampled batch segregated and unavailable for sale until it passes.

Reading and recording the result

• Match the result to the tracking-system batch or package identifier before using it for release or labeling.
• Confirm every required test appears for the product type and state, and do not read a potency result as proof that contaminant tests passed.
• Review units, quantitation limits, pass or fail fields, preparation notes, and qualifiers before entering values anywhere else.
• Confirm the tested formulation and package configuration match what will be sold.
• In Missouri, confirm results were filed in the tracking system within ten days of sample collection and within 24 hours of the analysis being completed.

Label data and the November 1, 2026 Illinois fields

• Calculate label cannabinoid values from the correct report and check rounding and unit conversions before artwork approval.
• For Illinois labels, confirm the report supports the Total THC value, the THC, THCA, CBD, and CBDA values, and any other cannabinoid above the threshold described in the Spring 2026 bulletin.
• Record which report and formula produced each label value, and update the product registration when ingredients change.

Failures, holds, and retests

• Follow a written decision tree for failed results that names who freezes inventory, who reviews the report, and which path the rule allows.
• In Illinois, recall and dispose of a batch that fails the pesticide residue test, and send other failures only to a CO2 or solvent-based extract that is retested.
• In Missouri, notify the Department of the chosen path within 15 days of a failure and reanalyze, remediate, or destroy the product within three months.
• Do not replace a failed or passing result with a different certificate outside the process the rule allows.

Records and downstream use

• Provide each dispensary with the results for every batch it buys, and keep the original laboratory report, not a reformatted marketing certificate.
• Tie every correction to an audit trail showing who changed the record, why, and which report supports it.
• Audit a sample of released inventory against reports and tracking records on a documented schedule.

Illinois: sampling and release

Each batch must be made available at the cultivation center, craft grower, or infuser for an approved laboratory employee to select a random sample immediately before manufacturing, natural processing, or packaging for sale to a dispensary. Required tests cover microbiological contaminants, mycotoxins, pesticide active ingredients, residual solvents, heavy metals, and an active ingredient analysis. The sample must be 7 to 15 grams drawn from different parts of the batch, cured flower batches may not exceed 20 pounds, and each final batch must pass all testing before it is offered for sale. A pesticide failure requires recall and disposal of the entire batch. Other failures allow the batch to be used for a CO2 or solvent-based extract, which must pass all required tests. Laboratories file microbiological, mycotoxin, heavy metal, and pesticide failures with the Department when they send results to the producer, keep results for at least five years, and return or dispose of tested cannabis. Producers must give dispensaries the results for each batch they purchase, and dispensaries must make them available to purchasers on request (8 Ill. Adm. Code 1300.700). The section was amended effective May 1, 2026, so confirm any internal procedure written before that date.

The Illinois label fields due November 1, 2026

The Department’s Spring 2026 bulletin makes the testing laboratory the source of the new Total THC figure on Illinois labels and requires more cannabinoids to be listed from the certificate. For the testing program, that means more values move from the report to the artwork, so each one needs a recorded source and a second check. The adopted text of the labeling section should be confirmed with the Department, because the online version we reviewed has not been updated. Our guide to advertising and packaging lists the new fields, and our explainer on total THC and delta-9 explains the calculation.

Missouri: lots, reporting, and remediation

Missouri requires testing licensees to test every lot of marijuana product, other than seeds and plants, before sale. Sampling is done by the testing licensee at the harvest or process lot level. Dry, unprocessed marijuana is sampled in pools of up to 15 pounds covering at least 0.5 percent of the harvest lot, and extracts, cartridges, prerolls, and infused products have their own sample counts. Results must be filed in the tracking system within ten days of collection and 24 hours of analysis, before or at the same time the originating facility is told. A passing lot may not be retested to replace its results without written approval. The profile covers eight cannabinoids, and contaminant limits cover microbials, pesticides, heavy metals, residual solvents, water activity and moisture, and foreign matter. Failed product goes on administrative hold and must be reanalyzed, remediated, or destroyed within three months, heavy-metal failures may not be remediated, and sampling manipulation can bring fines of up to $100,000 (19 CSR 100-1.110).

The Division’s June 12, 2026 guidance on bulk sampling adds practical limits. Harvest lots may be bulk sampled only after all processing is finished. Products sold in their method of administration, such as vape cartridges, may not be bulk tested. Infused products sampled in bulk must be divided into sale units before testing, with the unit count on the chain of custody. The guidance also notes that recalls run by test tag number, so using one set of results for several products can widen a recall.

The Missouri text above was read from a reproduction current through June 2025. The Department’s rules took effect in their current form on May 30, 2026, so confirm each figure against the current rule.

Handoffs are where errors enter

Many testing errors happen after the laboratory has done its work correctly. A report arrives under a filename without the package number, someone copies potency into a spreadsheet, artwork uses an old result, or a distributor receives a certificate that drops the qualifiers. Treat each handoff as a controlled data transfer. Define the authoritative identifier, require a second check of manually entered values, and block label release until the batch, tracking identifier, laboratory report, and packaging record agree. When the method, laboratory, formulation, or legal requirement changes, review the label templates and release criteria that depend on it.