April 20, 2026

Just how to Check out a COA for THCA Products: A Newbie's Overview

A good Certificate of Analysis is the difference between buying with confidence and crossing your fingers. If you are shopping for THCA flower, vapes, concentrates, or infused edibles, the COA tells you what is in the product, what is not, and how sure the lab is about both. The paper looks technical at first glance, full of abbreviations and numbers, but once you know how the page is organized you can scan it in under two minutes and catch the signal from the noise.

I have sat across from cultivators arguing why their numbers should round up, and from retailers who were blindsided when a product failed a state audit. The common thread is always the COA. When you learn to read one, you spot problems early, you match your expectations to the chemistry, and you avoid marketing that tries to turn lab jargon into hype.

What a COA is, and why THCA makes it tricky

A COA is a lab report. It documents the sample, the testing methods, the results, and the uncertainties. For THCA products, there is a specific wrinkle: THCA is the acidic, non-intoxicating precursor to delta‑9 THC. Heat, time, and light convert THCA to THC, a process called decarboxylation. Flower and many concentrates start with mostly THCA. When you smoke or vape, most of that THCA converts to THC. Edibles and distillates are often already decarboxylated. This means a THCA COA usually lists both THCA and THC, and the meaningful number for expected psychoactive effect depends on whether the product will be heated.

This detail also touches legal compliance. Federal hemp rules set a 0.3 percent delta‑9 THC limit by dry weight. Many jurisdictions look at delta‑9 THC only. Others consider total THC, which is the amount of THC present plus the amount that will form when THCA decarboxylates. Understanding which standard applies to you changes the interpretation of the same sheet of paper.

The sections of a solid COA

Some labs add their own branding or arrangement, but the backbone is fairly consistent. A complete COA for a THCA product should include:

A header with the lab name, accreditation status, report date, and report number
Client and sample information, including product type, batch or lot number, strain or SKU, and sample dates
Methodology and instrumentation for each panel tested, such as HPLC for cannabinoids
Results by panel, often cannabinoids, terpenes, moisture or water activity, heavy metals, pesticides, residual solvents, mycotoxins, and microbials
Quality notes such as limits of detection (LOD), limits of quantification (LOQ), and measurement uncertainty
Signatures and reviewer info, sometimes with a QR code or URL for verification

If any of these are missing, ask why. Sometimes a COA is split into multiple pages or separate reports by panel. Perfectly fine, as long as they clearly link to the same batch.

How to verify it is a legitimate, traceable report

The first pass is about authenticity. Look for an ISO/IEC 17025 accreditation statement, with the accreditation body’s mark. Labs earn this by proving competence and quality systems. An unaccredited lab can still produce accurate work, but accreditation removes guesswork about calibration, documentation, and method validation. Next, scan the QR code or use the provided URL to pull the report directly from the lab portal. If the PDF you were sent cannot be found online, or the numbers do not match, do not buy until that is resolved.

Check the sample and batch details. The batch or lot number on the COA should match the label on the jar, bag, cartridge, or box. If the vendor says the COA is “for the same product, different batch,” you cannot assume equivalence. Batch-to-batch variation is real. For THCA flower, a 1 to 3 percent swing in potency is common. For a gummy line, batches can vary a few milligrams per piece if process control slips. Chain of custody and sample date matter too. A report dated long after the product label date, or a report missing a sampling date, can be a sign the wrong material was tested or the product aged beyond the tested state.

Finally, find the signature block. There should be a named analyst or technical reviewer who signed off. Digital signatures are common. Labs also include a statement of conditions, such as sample was received intact, or analysis is valid only for the items tested. These guardrails tell you the lab is not hand waving.

Cannabinoid panel: reading THCA the right way

Most shoppers jump to the cannabinoid table first. This is where the product’s potency lives. You will see analytes like THCA, delta‑9 THC, delta‑8 THC, CBDA, CBD, CBG, and sometimes minor cannabinoids like THCV. The results are usually expressed as percent by weight for flower and solids, milligrams per gram for concentrates, and milligrams per milliliter for liquids. Some labs show both percent and mg/g side by side, since 1 percent by weight is equal to 10 mg/g for most solids.

For THCA products, focus on three things:

THCA percentage or mg/g
Delta‑9 THC percentage or mg/g
Total THC, which many labs calculate using the stoichiometric factor 0.877

Why 0.877? When THCA loses a carboxyl group during decarboxylation, its molecular weight drops. The conversion factor accounts for that mass change. In practice, total THC is often calculated as delta‑9 THC plus 0.877 times THCA. If the lab does not show a total THC value, you can estimate it quickly. For example, if a flower shows 0.2 percent delta‑9 THC and 21.0 percent THCA, total THC would be 0.2 + 0.877 × 21.0, or about 18.6 percent. That lines up with the experience of a potent flower when smoked, even though the measured delta‑9 THC on the raw flower is low.

Note the method in the panel header. Cannabinoids in plant material should be tested by high‑performance liquid chromatography, or HPLC. It separates the acidic and neutral cannabinoids without heating the sample, so THCA stays THCA during analysis. Gas chromatography heats the sample and can convert THCA into THC inside the instrument, which inflates the THC number and erases the THCA number. A COA that reports cannabinoids from GC on raw flower or live resin is a mismatch and hard to interpret.

A quick math story from the counter

A customer once brought me a bag labeled 25 percent THCA with a COA to match. The bag also said compliant delta‑9 THC below 0.3 percent. They asked if it would feel weak. We did the math together. The COA showed 0.24 percent delta‑9 THC and 25.1 percent THCA by HPLC. Using the factor, total THC was roughly 22.2 percent. That is robust. The 0.3 percent delta‑9 THC line matters for legality in some jurisdictions, but it does not predict effect when the product is heated. Once the customer saw the numbers side by side, the label made sense.

Dry weight basis, moisture, and why flower rarely reads 0.3 percent delta‑9 THC after smoking

A separate twist in hemp compliance is the dry weight basis. Flower has moisture. If you express an analyte’s percentage on an as‑received basis, moisture dilutes the math. Express it on a dry basis and the numbers go up. COAs should clearly state whether cannabinoids are reported on a dry weight basis. Labs often report water activity or moisture alongside cannabinoids for this reason.

Water activity, noted as aw, tells you how freely water moves in the material. For flower, an aw between about 0.55 and 0.65 is typical for stable storage. Above roughly 0.70, microbes have a party. Moisture percentage varies more widely with curing style, but a range between 8 and 12 percent is common. If the COA shows unreasonably low moisture, ask how the sample was prepared. If the flower’s moisture drops over time on the shelf, the as‑received potency may go up a hair, but the dry basis potency should stay consistent. This matters for legal calculations and for predicting potency after decarbing.

Units matter: mg/g, mg/mL, percent, ppm, and CFU

Potency numbers will hop between units depending on the matrix. Flower is often percent by weight. Concentrates may be mg/g. Vape liquids are sometimes mg/mL. Edible serving sizes are mg per piece. Metals and pesticides go to parts per million or parts per billion. Microbial results use colony forming units per gram, CFU/g.

Keep a mental converter: for solids and semi‑solids, 1 percent equals 10 mg/g. A gram of 70 percent THCA diamonds contains about 700 mg of THCA. If that product is dabbed or vaped, expected THC delivery depends on decarb completeness and device efficiency. Not all THCA converts, and not all vapor is inhaled. In the lab, total THC is a theoretical max based on chemistry. In the real world, bioavailability trims it down.

Legality checks and the nuance of total THC vs delta‑9 THC

Different states and sales channels apply different rules. A typical legal quick check is:

If you are in a jurisdiction that follows the federal hemp rule, confirm delta‑9 THC is at or below 0.3 percent on a dry weight basis for raw material. For edibles, the cap might be set per serving, not by weight percent.
If your state uses total THC for compliance, compute or confirm the lab’s total THC value. For THCA flower, total THC usually governs legality.
Look at measurement uncertainty. If delta‑9 THC reads 0.29 percent with an uncertainty of ±0.05 percent, it is not safely below the line. Regulators can decide how to handle that overlap.
Check that the test panel list includes any compliance panels required by the jurisdiction, not only potency.

A note about labels that market THCA flower as hemp because delta‑9 THC alone is below 0.3 percent: many regulators now look at total THC. Retailers have faced product seizures and fines for ignoring this drift. Before selling or shipping, align the COA math with your local standard.

Residual solvents, pesticides, heavy metals, and microbials

Potency is not the only story. A clean COA also covers safety contaminant panels. Which panels apply depends on the material.

Pesticides and heavy metals matter for any plant material. The plant uptakes metals from soil. Commonly reported metals include lead, arsenic, cadmium, and mercury. Hemp grown in contaminated soil can exceed limits. Pesticide screens cover dozens of compounds with strict limits. Fails do happen, and they are rarely fixable. If a COA shows levels just below the action limits, that is not automatically a problem, but it is a wake‑up call to monitor the supplier.

Residual solvents and volatiles apply when the product has been extracted or processed with solvents. Hydrocarbon extracts should be screened for butane, propane, hexane, and related solvents. Ethanol extracts should be screened for ethanol and other alcohols. CO2 extractions may still use ethanol during winterization, so do not assume solvent free without a test. The action limits are measured in ppm. A report line that reads ND, not detected, means the analyte was below the lab’s detection limit, not necessarily zero. The LOD and LOQ values printed on the COA tell you how low that limit sits.

Microbial and mycotoxin panels matter for flower and edibles. Mold and bacteria can proliferate in poorly cured or stored flower. Aflatoxins and ochratoxin A are serious risks. For gummies and chocolates, water activity is often low enough to control microbial growth, but process hygiene still counts. Labs report microbials as CFU/g with limits per category. If the COA flags a fail in any microbial category, consider the product unsafe, even if potency looks great.

Terpene data: helpful context, not a potency score

Many COAs now include a terpene panel. This is useful when you want to understand aroma, flavor, and part of the effect profile. A flower with 2 percent total terpenes will often smell and taste richer than one with 0.5 percent. Dominant terpenes like myrcene, limonene, and beta‑caryophyllene help you compare across batches and cultivars. For THCA concentrates, terpene levels swing more widely. A live resin might show 4 to 8 percent total terpenes, while a high purity THCA diamond fraction may be under 1 percent on its own, with separate terp sauce blended later. Do not treat terpenes as a scoreboard. They are one lens, Cheefbotanicals not the whole picture.

The two numbers most beginners misread: LOD/LOQ and uncertainty

Two small lines on the COA carry more weight than they seem.

Limit of detection is the lowest amount the method can confidently detect as present. Limit of quantification is the lowest amount the method can reliably measure with precision. If a pesticide is listed as ND with an LOQ of 10 ppb, that means if it is present, it is below 10 ppb. If your action limit is 100 ppb, that is good enough. If your action limit is 5 ppb, the method cannot confirm compliance.

Measurement uncertainty tells you how much wiggle room the result has. Every measurement has error. A cannabinoid reading of 21.5 percent with a ±1.0 percent uncertainty means the true value is expected to lie within 20.5 to 22.5 percent most of the time. For legal cutoffs, uncertainty matters. For consumer experience, a 1 percent swing on flower is not dramatic, but it is real when you compare jars side by side.

Common red flags that save you from bad buys

Over the years, I have learned to trust my nose for off COAs. A few patterns repeat:

The lab is unaccredited and the report lacks method details or LOD/LOQ. Not always a deal breaker, but a slow down and ask questions moment.
The numbers are too neat. Exact whole numbers for multiple cannabinoids, totals that round to 100 percent, or copy‑pasted lines across batches. Real data has noise and decimals.
THCA or THC values that exceed 1000 mg/g or totals that imply more than 100 percent by weight. Impossible on a mass basis.
A cannabinoid panel is HPLC for flower on one page, then a different page for the same batch shows GC results with different values. Mixed methods without explanation can mask a problem.
QR code links to a generic landing page, not the specific result, or the linked report does not match the provided PDF.

If you see one of these, slow down and ask for clarification from the vendor or the lab. Good suppliers welcome the questions.

Serving size math: from lab sheet to pocket dose

A frequent point of confusion is converting the COA into a serving. Here is a practical way to do it for different product types:

Flower and prerolls: Use total THC to estimate per gram potency for inhalation. If the flower shows 20 percent total THC, that is about 200 mg THC potential per gram of flower. Smoking is not 100 percent efficient. Depending on the device and technique, delivered THC might land around 30 to 50 percent of that number. So a half gram joint could deliver roughly 30 to 50 mg of THC to the bloodstream, often less for casual puffing.

Concentrates: If a THCA diamond jar reads 90 percent THCA and 1 percent THC, total THC potential is roughly 0.877 × 90 + 1, or 80.9 percent. A typical small dab might be 30 mg of material, which holds about 24 mg THC potential. Again, device and technique affect delivery.

Vape carts: Units are often labeled in mg/mL. If the COA says 700 mg/mL total THC potential and the cart is 1 mL, the cart holds about 700 mg potential. Each 3 second draw from a typical pen vaporizes 3 to 5 mg of liquid, delivering a few milligrams of THC. Variability is high across hardware.

Edibles: Look for mg per piece or mg per serving. If the COA shows 10 mg THC per gummy, that should match the label. THCA edibles are unusual, since most edibles are decarbed during cooking. If you see THCA listed in an edible, ask the manufacturer how they ensure consistent conversion upon consumption.

These estimates help set expectations. Keep in mind tolerance, diet, and metabolism change the experience.

Step‑by‑step way to read a THCA COA without missing something important

Verify the lab and the link. Check ISO 17025 status and pull the report from the lab portal via QR or URL.
Match the batch. Confirm product type, lot number, and sample dates align with what you are buying or selling.
Read the cannabinoid panel. Note THCA, delta‑9 THC, total THC, units, method, and whether results are dry weight.
Check safety panels relevant to the product. Metals and pesticides for plant material, solvents for extracts, microbes for anything ingestible or smokable.
Scan uncertainty, LOD, and LOQ. If a number hugs a legal limit, decide whether the margin is acceptable in your context.

Keep this flow, and you will spot most problems in seconds.

Case examples from the field

A retailer friend called after a surprise inspection. Their THCA flower line had COAs that showed under 0.3 percent delta‑9 THC and around 22 percent THCA. They thought they were safe to ship across state lines. The state auditor applied a total THC rule. The COAs were technically accurate, but the interpretation was wrong for that jurisdiction. They had to quarantine the inventory and rework their supply chain for compliant channels. The lesson was not about chemistry, it was about reading the legal column alongside the lab column.

Another time, a small extractor sent me a COA with residual solvent fails. Butane measured at 540 ppm against a 500 ppm action limit. They argued that the batch had been purged further since sampling. The lab would not change the report, and rightly so. We resubmitted a new sample after extended purging. The retest came back at 120 ppm. Both COAs were honest snapshots in time. Without understanding LOD, LOQ, and the fact that ND is not the only acceptable safe result, the team might have tossed salvageable material or, worse, shipped the failed batch as‑is.

A grower brought me a terpene panel to explain why a strain that always sold well was lagging. The new batch showed 0.7 percent total terpenes vs 2.1 percent typical. Potency was similar. The batch smelled flat. It had been over‑dried, water activity at 0.40. The COA told the story in numbers. They adjusted the cure room humidity and the next harvest climbed back over 1.8 percent. Sales followed.

Practical notes for different product types

THCA flower: Look for robust THCA with low delta‑9 THC if legality hinges on delta‑9 alone, but compute total THC for expected effect. Moisture and water activity should be present. Pesticides and metals are non‑negotiable. Microbial screens are wise, especially in humid climates or for long shipping.

THCA concentrates and diamonds: The cannabinoid panel should specify high THCA content with low delta‑9 THC for raw fractions, unless it is a decarbed product. Solvent screens must be present for hydrocarbon extracts. Terpene data is helpful if a sauce fraction is included. Watch for lab methods that maintain THCA integrity during analysis.

Vapes: Units should be mg/mL for potency. Viscosity modifiers should be absent unless disclosed and tested for safety. Residual solvent and metals testing is essential, along with leachables testing from hardware if available, though that often appears in separate validation documents rather than every batch COA.

Edibles: Potency should be per piece and per package. Homogeneity testing adds confidence that each unit matches the label. Microbials and mycotoxins matter. For products that claim THCA content, ask for method and stability data, since stomach acid and body temperature are not reliable decarb systems for consistent results.

How to avoid being misled by labels and marketing

Labels have limited space and lots of incentive to round up. A jar might shout 30 percent THCA because one bud tested high, while the batch average on the COA is 26 percent. Labs sometimes provide both raw data and batch averages. A fair label uses averages and states whether values are typical or maximum. Some brands print the QR code on the label so you can verify. That is a good sign.

Beware of terms like solventless used on products that were winterized with ethanol, or pesticide free without a pesticide panel. For THCA specifically, watch for made‑up conversion claims, like 100 percent conversion on every hit, or total THC above 1000 mg/g after conversion. Chemistry sets the ceiling.

Storage, stability, and expiration dates

THCA does not sit still. Over time, even at room temperature, some THCA converts to THC. Oxygen and light also drive degradation to CBN and other byproducts. A COA is a snapshot on the sample date. If you are reading a year‑old COA for a jar that has lived on a sunny shelf, expect drift. Not all COAs list an expiration date, but they should at least show sample and report dates. For retail, a potency recheck at 6 to 12 months for slow‑moving inventory is smart, especially in warmer climates.

For consumers, store flower in an airtight container in a dark, cool place. Keep vapes away from heat. THCA concentrates keep well cold. These steps preserve the profile the COA described.

A concise checklist you can use at the counter

Confirm the lab’s ISO 17025 accreditation and verify the report via QR or URL.
Match product type, batch or lot number, and dates between label and COA.
Read cannabinoids with units, check method is HPLC for raw plant THCA, and compute total THC if not shown.
Ensure required safety panels are present and pass: pesticides, metals, solvents, microbials as applicable.
Note measurement uncertainty, LOD, LOQ, and whether values are dry weight. Decide if margins near limits are acceptable.

Keep this in your notes app. It covers 90 percent of decisions you will face.

What to ask a vendor or lab when something looks off

If the COA raises questions, ask direct, practical ones. Request the raw data or a method summary if the analyte list seems odd. Clarify whether results are dry weight or as received. If the product is marketed across jurisdictions, ask how they determine compliance where total THC rules apply. For concentrates, ask about solvent recovery and purge conditions, not just the pass or fail. Good partners will have the answers ready, and they will not take offense.

Final thoughts from the lab bench

Reading a COA for THCA products is part chemistry, part regulation, and part common sense. You do not need to memorize every acronym. You need to know where to look and how to connect the lines. THCA shows up as potential on paper and as effect when heated. Total THC bridges that gap. Safety panels protect you from invisible problems. Uncertainty and detection limits put honest boundaries around the numbers.

Once you practice a few times, the page stops being a wall of text and turns into a map. You glance at the header, slide to the cannabinoids, skim the safety blocks, and if everything lines up, you buy with confidence. If it does not, you walk away or ask better questions. That habit protects your wallet, your customers, and your reputation, which is the whole point of testing in the first place.