How to read a cannabis resin COA: a section-by-section guide for buyers

What a certificate of analysis for cannabis resin tells you, and what it doesn't: the header, cannabinoids and total THC, pesticide residues, aflatoxins and microbiology, explained with an illustrative example.

For a licensed buyer, the certificate of analysis (COA) is the most important document in a resin purchase. It tells you what you are importing, whether it meets your market's limits and whether the batch you receive is the batch that was tested. Yet COAs vary widely in layout, depth and language, and the details that matter most are often in the footnotes.

This guide takes a cannabis resin COA section by section, the way an importer's quality team reads one: the header, cannabinoids, pesticide residues, mycotoxins, microbiology and the fine print.

1. Start with the header: what was tested, and when

Before looking at any number, read the header. It tells you which sample the results belong to, and a COA is only as useful as that link.

FieldExample (illustrative)What to check
ClientThe producer's company nameMatches the legal entity on the producer's licences
Sample typeCannabis resinMatches the product you are buying
IdentificationLot: EXAMPLE-001The lot number matches your delivery
Sampling placeProduction unitWhere the sample was taken
Sampling dateDay 1How old the data is relative to your shipment
Received / analysedDay 2Time between sampling and analysis
Report issuedDay 4The report date, not the sampling date
Sampled byThe laboratoryWho took the sample: the lab or the producer

Three questions matter most here. Is the lot number the lot you are buying? A COA for a different lot tells you about the producer's typical quality, not about your goods. How recent is it? Cannabinoid levels can change during storage, so an older COA is less representative. Who took the sample? Laboratory sampling or a documented sampling plan gives more assurance than a sample picked and sent by the seller. The report's fine print usually says who was responsible.

2. Cannabinoids: THCA, THC and the "total" figures

The cannabinoid section is the one most buyers read first. Resin is usually analysed by liquid chromatography (HPLC), often with UV detection (HPLC-UV). An illustrative result might look like this:

CannabinoidExample result (illustrative)
CBDA1.00%
CBD3.00%
Total CBD3.88%
THCA20.00%
THC2.00%
Total THC19.54%

Acid and neutral forms. Cannabis makes cannabinoids in their acid forms, THCA and CBDA. Heat converts them to THC and CBD (decarboxylation), and so does time, slowly. HPLC does not heat the sample, so it measures the acid and neutral forms separately. That is why a good resin COA lists both.

How "total" is calculated. When THCA loses its carboxyl group it also loses mass, so 1 g of THCA gives about 0.877 g of THC. The standard formula is:

In the example: 2.00 + 0.877 × 20.00 = 19.54%. Total CBD works the same way: 3.00 + 0.877 × 1.00 = 3.88%. Checking this arithmetic on a real COA takes ten seconds and confirms the lab used the standard conversion.

What to ask. Is the percentage weight per weight (w/w)? What is the measurement uncertainty? Many reports print neither, and conclusions are often stated without accounting for uncertainty. If your contract has a potency threshold, agree how uncertainty is handled. Also note which cannabinoids were tested: if your specification needs CBG, CBN or other minor cannabinoids, ask for them explicitly.

3. Pesticide residues: long lists and the limit of quantification

Pesticide screening is where resin COAs differ most, because the number of substances tested can range from a handful to several hundred. Broad screens typically run on two instruments:

  • GC-MS/MS (gas chromatography with tandem mass spectrometry)
  • LC-MS/MS (liquid chromatography with tandem mass spectrometry)

In Europe, laboratories often base their method on NF EN 15662, the European standard for multi-residue pesticide analysis, and assess results against EU maximum residue limits (MRLs).

Three things to read carefully:

  1. The limit of quantification (LQ). A typical LQ for a broad screen is around 0.01 mg/kg per substance, sometimes lower for substances with strict limits. A result "< LQ" means the substance, if present at all, is below what the method can reliably measure.
  2. The scope of the list. A clean result on 20 substances is not the same as a clean result on several hundred. Compare the list with the substances your market regulates.
  3. The conclusion. The strongest outcome of a residue screen is a statement that all residues tested are below the limits of quantification.

A detail worth knowing: long pesticide tables often list only the substance and its LQ, with the overall result in a single conclusion line. That is normal. It means nothing was quantified, so there is no per-substance value to print.

4. Mycotoxins: aflatoxins and what may be missing

Resin comes from dried plant material, and poorly dried or badly stored material can grow moulds that produce mycotoxins. Laboratories commonly test the four regulated aflatoxins by UPLC-MS/MS or a similar method:

AflatoxinExample result (illustrative)Example LQ
B1< LQ0.5 µg/kg
B2< LQ0.5 µg/kg
G1< LQ0.5 µg/kg
G2< LQ0.5 µg/kg
Total aflatoxins< LQ2.0 µg/kg

Note the units: µg/kg, a thousand times smaller than the mg/kg used for pesticides. Look too for a footnote on recovery. Many labs report results "not corrected for recovery", meaning they report what they measured without adjusting for the share of analyte lost during sample preparation. That is common practice, but it matters when results are close to a limit.

Aflatoxins are not the only mycotoxins. Depending on your market, your specification may also require ochratoxin A. If it is not on the COA, ask for it before the contract rather than after the goods arrive.

5. Microbiology: counts, absences and pharmacopoeia criteria

For markets that apply pharmaceutical expectations, microbiology is often tested with European Pharmacopoeia methods: 2.6.12 for microbial counts and 2.6.31 for specified micro-organisms in herbal products. Results are assessed against stated criteria:

TestMethodExample criterionExample result (illustrative)
Total aerobic microbial countPh. Eur. 2.6.12≤ 10⁵ CFU/g1.0 × 10³ CFU/g
Total yeast and mould countPh. Eur. 2.6.12≤ 10⁴ CFU/g1.0 × 10² CFU/g
Escherichia coliPh. Eur. 2.6.31Absence in 1 gAbsence
SalmonellaPh. Eur. 2.6.31Absence in 25 gAbsence
Bile-tolerant gram-negative bacteriaPh. Eur. 2.6.31< 10⁴ MPN/g< 10 MPN/g

Reading the numbers. CFU means colony-forming units. "1.0 × 10³" (written 1,0.10³ on some French-language reports) is 1,000 CFU per gram. MPN is the "most probable number", a statistical count used for some tests. Absence tests are pass/fail, and the quantity matters: Salmonella is usually tested in 25 g, E. coli in 1 g. The criteria themselves come from your market's requirements or the pharmacopoeia edition cited, so check that they match your specification.

A detail that signals a careful lab. Good reports state that the method's suitability, including neutralisation of any antimicrobial activity in the sample, was tested on the product. Plant materials can suppress microbial growth in the test itself, which would make counts look better than they are. Validation rules that out.

6. The fine print that changes how you read everything else

Laboratory reports carry standard statements that are easy to skip and important to understand. Look for these:

  • Results relate only to the items tested, as received by the laboratory. The COA does not certify an entire production run unless the sampling plan was designed for that.
  • The report may only be reproduced in full. Be wary of cropped or partial COAs.
  • Accreditation. A conclusion may be covered by the laboratory's accreditation only if every parameter is accredited. Ask which parameters are within the accredited scope.
  • Uncertainty. Conclusions are often stated without taking measurement uncertainty into account.
  • Sample retention. Labs keep samples for a limited time after analysis, often a couple of weeks, so any retest has to be requested quickly.

None of these is a warning sign; they are how accredited laboratories normally report. But they define what the COA does and doesn't prove.

7. What a COA may not cover

A COA answers only the questions that were asked. Depending on your market, product and use, your specification may also require:

  • Moisture or water activity, which matters for storage stability and microbial risk
  • Heavy metals (for example lead, cadmium, mercury and arsenic)
  • Residual solvents, mainly relevant for solvent-based extracts; solventless resins are made without solvents
  • Terpene profile, when aroma is part of the specification
  • Additional mycotoxins, such as ochratoxin A
  • Foreign matter and appearance

The practical rule: agree the test panel, the laboratory and the acceptance limits in the contract, before the batch is produced or reserved.

A buyer's COA checklist

  1. The lot number on the COA matches the lot on your delivery documents.
  2. The sampling date is recent enough for your purpose, and you know who took the sample.
  3. Cannabinoids include acid and neutral forms, and total THC and CBD check out arithmetically.
  4. Units are clear: % (ask whether w/w), mg/kg for pesticides, µg/kg for mycotoxins, CFU/g for microbiology.
  5. The pesticide scope covers what your market regulates, with LQs at or below your limits.
  6. Aflatoxins, and any other mycotoxins your market requires, are reported.
  7. Microbiology follows a recognised method, such as the European Pharmacopoeia, with stated criteria.
  8. You know which parameters fall within the lab's accredited scope.
  9. Anything your specification requires but the COA lacks has been requested in writing.
  10. The report is complete, signed and unaltered.

How we handle COAs

Every batch of our solventless resin, from Dry Resin to Static Resin and Powder Resin, has a certificate of analysis. Cannabinoids are profiled in our on-site HPLC laboratory, every batch has a terpene profile, and contaminants are tested by independent laboratories. Full reports, together with our ANRAC authorisations and our LSQA WHO-GACP declaration, are shared with approved buyers in the portal. For the licences and test methods behind them, see Quality & Compliance. To evaluate a specific product, use the Request a COA button, or apply as a licensed buyer.

Why is total THC higher than THC on a resin COA?

Most of the THC in resin is present as THCA, the acid form. Total THC converts THCA to its THC equivalent (multiplying by 0.877) and adds the THC already present. With illustrative values of 2.0% THC and 20.0% THCA, total THC is 2.0 + 0.877 × 20.0 = 19.54%.

What does "< LQ" mean?

The result is below the limit of quantification: the lowest concentration the method can measure reliably. It does not prove the substance is completely absent, only that any amount present is below that limit.

Is one COA valid for every delivery?

No. A COA describes the sample that was tested, and accredited laboratories state that results relate only to that sample. Match the lot number on the COA to the lot you receive.

How do I get a COA for your products?

Approved buyers find full laboratory reports in the buyer portal. You can also request a COA for a specific product with the Request a COA button on this page or on any product page.