A COA Can Prove a Lot. It Can't Prove Everything.

A QR code is not quality control.

It is a doorway.

What matters is what is on the other side.

A real Certificate of Analysis can be one of the most useful pieces of information attached to a hemp product. It can show you what a laboratory measured, what sample was tested, when it was tested, and—when the paperwork is being handled correctly—which batch that sample is supposed to represent.

That is valuable.

It is also narrower than the way the industry sometimes talks about it.

“Lab tested” gets used like a universal quality badge. Scan the code. See a PDF. Case closed.

Except a laboratory report is not the product itself. It is evidence produced from a sample of the product.

Start With the Most Important Question: What Was Actually Tested?

Not every COA tests for the same things.

One report might contain a cannabinoid potency panel. Another might include cannabinoids, pesticides, heavy metals, residual solvents, microbiology, mycotoxins, terpenes and moisture-related measurements.

Those are different reports answering different questions.

If a cannabinoid panel says a flower sample tested at a certain concentration of THCA, that is evidence about the cannabinoid concentration measured in that sample.

It does not automatically prove that the same sample passed a pesticide screen.

If pesticides were not part of the analysis, the cannabinoid result cannot answer that question for you.

The distinction sounds obvious when you spell it out.

It gets lost constantly when the entire report is reduced to two words:

LAB TESTED.

The Laboratory Does Not Test the Entire Harvest

This is where yesterday's article becomes important.

We explained why the same strain can produce different flower from one batch to another.

Variation does not magically disappear inside one harvest either.

Cannabis flower is biological material. Different parts of the same plant can vary. Different plants in the same cultivar can vary. Different locations inside a harvest lot can vary.

That creates a basic testing problem.

A laboratory cannot grind up every flower in a sellable harvest and then hand the product back.

Testing requires sampling.

The goal is to collect a sample that represents the larger batch closely enough that the resulting measurements are meaningful.

There is an entire ASTM standard devoted to exactly this issue. ASTM D8334/D8334M covers sampling post-harvest cannabis and hemp flower for laboratory analysis. The standard specifically notes that cannabis material can vary between different portions of the same plant and between plants within the same cultivar.

That is why representative sampling matters.

If the laboratory receives a sample that does not represent the batch, extremely precise measurement of that sample still does not magically make it representative.

Good analysis starts before the sample ever reaches an instrument.

Representative Does Not Mean Every Nug Was Individually Tested

This distinction matters because consumers understandably read a batch COA as though the laboratory tested the exact piece sitting in their jar.

Usually, that is not what happened.

The laboratory analyzed material sampled from the larger batch.

If sampling and batch controls were handled appropriately, the report is intended to characterize that batch.

That is normal analytical testing. It is not a loophole.

It is also why the words batch specific matter so much.

Imagine buying a new harvest of a strain and scanning a QR code that sends you to a beautiful laboratory report from six months ago.

The strain name matches.

The brand matches.

The product category matches.

The batch does not.

At that point, the PDF may be legitimate laboratory work. It just is not evidence about the harvest currently sitting in your hand.

Then the Laboratory Has to Measure It Correctly

Sampling is only the first half of the measurement chain.

Once the material reaches the laboratory, somebody still has to prepare it, extract the compounds being measured, run an analytical method, calibrate the instrumentation, interpret the data and report the result.

Those steps matter.

The National Institute of Standards and Technology has spent years developing cannabis laboratory quality-assurance tools for exactly this reason.

NIST's Cannabis Laboratory Quality Assurance Program was designed to help laboratories compare their measurements against accepted values and against the broader laboratory community. NIST has also developed hemp reference materials containing carefully characterized cannabinoid and toxic-element concentrations that laboratories can use to check whether their analytical methods are producing accurate results.

Think of a reference material as a known answer.

If a laboratory measures a material whose composition has already been carefully characterized and its result lands where it should, that gives the lab evidence that its measurement system is behaving properly.

If it does not, something needs investigation.

That does not mean cannabis labs are inherently untrustworthy.

It means chemistry is measurement, and measurement requires quality control.

A machine does not become infallible because the final result gets exported as a PDF.

Even Something as Boring as Moisture Can Change the Measurement Problem

NIST's quality-assurance work offers a good example of how unglamorous laboratory details can matter.

In a 2025 cannabis laboratory quality-assurance exercise, participating laboratories reported a wide range of moisture results for the same hemp material while using several different drying methods.

NIST concluded that the range demonstrated a need for consistent drying methods to produce accurate and precise moisture measurements.

Nobody buys hemp because they are excited about analytical moisture methodology.

But moisture and dry-weight calculations can matter to reported concentrations.

This is the larger point.

Behind a clean number on a COA is a chain of sampling, preparation, methodology, calibration and interpretation that has to work correctly for the number to deserve your confidence.

A COA Can Tell You Chemistry. It Cannot Smell the Jar for You.

Now we get to the other limit.

Some product characteristics simply are not the question the laboratory report was designed to answer.

A cannabinoid panel cannot tell you whether the trim job is good.

It cannot tell you whether the flower spent the last month baking in a hot warehouse.

It cannot tell you whether the jar seals properly.

It cannot tell you whether a disposable's coil will fail halfway through the oil.

It cannot tell you whether you will like the aroma.

It definitely cannot tell you exactly how you personally will feel after using the product.

Even a full terpene panel is still a chemical analysis—not a personalized prediction engine.

We covered that distinction in our guide to reading a hemp COA's terpene panel. The panel is extremely useful for understanding and comparing what is actually present. It should not be turned into another horoscope.

Testing Has an Expiration Problem Too

The laboratory measured the sample at a particular moment.

The product kept existing afterward.

Flower can continue changing during storage. Volatile compounds can be lost. Cannabinoids can degrade. Moisture can move in the wrong direction. Packaging and environmental conditions matter.

The COA does not continuously update itself as the product ages.

That is one reason the test date is useful context.

A result from the correct batch is meaningful.

But it still tells you what was measured in the tested sample at the time of analysis—not what happened during every day of storage after that.

The closer you get to treating a COA like a live telemetry feed from inside the package, the further you get from what the document actually is.

Four Things to Match Before You Trust the PDF

You do not need to audit the laboratory yourself every time you buy flower.

Start with four basic checks.

Then read the report for the question you are actually trying to answer.

Do not ask a potency panel whether the product passed a contaminant test that never appears on the page.

What Does “ND” Actually Mean?

You will often see ND somewhere on a laboratory report.

Usually it means not detected, but the exact reporting convention and threshold should be defined by the laboratory or the report legend.

It should not automatically be read as mathematical zero.

Analytical methods have limits to what they can reliably detect or quantify. A compound reported as not detected generally means the laboratory did not detect it at or above the relevant method/reporting threshold used for that analysis.

If that distinction matters for the result you are evaluating, check the report's definitions rather than guessing what the abbreviation means.

Use the COA for the Question It Can Actually Answer

The lesson here is not to trust laboratory testing less.

It is to use laboratory testing more intelligently.

A batch-specific COA can give you evidence that marketing copy never will.

It can tell you what cannabinoids were measured.

It can show you terpene concentrations when that panel was performed.

It can show contaminant results for the categories actually tested.

It can connect a package to a documented batch.

Those are not small things.

Just stop asking the document questions outside its jurisdiction.

Use the report for chemistry.

Use the batch number for traceability.

Use the test date for context.

Use your eyes and nose for physical condition.

Use good storage practices after purchase.

And use your own experience instead of expecting a PDF to predict yours.

That is not a limitation of good testing. It is what good testing actually looks like when we stop turning it into marketing.

---

FAQ

What does a hemp COA actually tell you?

A Certificate of Analysis reports the results of laboratory analyses performed on a submitted sample. Depending on the panels ordered, that can include cannabinoids, terpenes, pesticides, heavy metals, residual solvents, microbiology, mycotoxins and other measurements. Always check the report itself to see which analyses were actually performed.

Does a COA mean every product in the batch was individually tested?

No. Laboratory batch testing generally relies on representative samples collected from a larger batch. Appropriate sampling procedures are important because the tested material needs to accurately represent the batch the report is being associated with.

Does “ND” on a COA mean there is absolutely zero of that compound?

Not necessarily. ND generally means “not detected” under the laboratory's method and reporting criteria. Analytical methods have detection and reporting limits, so check the report legend or laboratory definitions when the distinction matters.

Can different laboratories produce different cannabinoid results?

Analytical measurements can be affected by methodology, calibration, sample preparation and other laboratory variables. Programs such as NIST's Cannabis Laboratory Quality Assurance Program exist specifically to help laboratories assess and improve measurement comparability and competence.

What should I check before trusting a hemp COA?

Match the product, batch or lot identifier, test date and analytical panels to the item you are evaluating. Then make sure you are using the report to answer only the questions covered by the analyses that were actually performed.

---

Sources & Further Reading

ASTM International. D8334/D8334M — Standard Practice for Sampling of Cannabis/Hemp Post-Harvest Batches for Laboratory Analyses. Best-practice standard addressing representative sampling of harvested cannabis/hemp inflorescence.

National Institute of Standards and Technology. NIST Tools for Cannabis Laboratory Quality Assurance. NIST program covering analytical methods, cannabis reference materials and laboratory quality-assurance tools.

Yarberry AJ, Phillips MM, Wilson WB. Cannabis Laboratory Quality Assurance Program: Exercise 2 Cannabinoid Final Report. NISTIR 8519. 2024. Interlaboratory quality-assurance exercise examining cannabinoid measurement comparability and laboratory competence.

National Institute of Standards and Technology. Reference Material 8210 Hemp Plant. 2024. Hemp reference material developed for cannabinoid, toxic-element and laboratory quality-control measurements.

Yarberry AJ, Barber C, Wood L, Wilson W. Cannabis Laboratory Quality Assurance Program: Exercise 3 Moisture Final Report. NISTIR 8574. 2025. Interlaboratory exercise illustrating variability in hemp moisture measurements produced using different drying methods.

---

These statements have not been evaluated by the Food and Drug Administration. These products are not intended to diagnose, treat, cure, or prevent any disease. For adults 21+ only. Individual experiences vary. Always review product labeling and verify applicable state and local laws before ordering or using hemp-derived products.