Third-Party Lab Tested: Why a Kava COA Matters More Than Any Label

"Third-party lab tested" is printed on a staggering number of kava products right now. It's on powders, on capsules, on shots, on cans. It costs a brand nothing to write, nobody polices the phrase, and in a lot of cases no independent lab ever opened the container.

The version of that claim that means something is a document. A certificate of analysis, tied to the specific lot in your hand, issued by a lab that has no financial reason to make the brand look good. I'm Pedro, co-founder of Kava Moon. We publish ours, including the parts that are less flattering, and this is the guide I wish I'd had when I first started reading these reports and had no idea which numbers mattered.

TL;DR

• A COA is a lab report for one batch of kava, not for a brand or a product line. If it isn't tied to a lot number, it isn't a quality record.
• It should answer three questions: identity (is this really Piper methysticum root), potency (how much of each kavalactone), and purity (heavy metals, microbials, pesticides, solvents, moisture).
• Kava's one serious safety episode in the West traced back to raw material fraud that testing catches: manufacturers using leaves and stem peelings instead of root.
• A COA is the only practical way to confirm kava is noble rather than tudei, because the proof lives in the flavokavain-to-kavalactone ratio, not in what the bag says.
• "Third-party tested" with no lab named and no document you can open is marketing. ISO/IEC 17025 accreditation plus a matching lot number is the real thing.
• Kava Moon publishes both our independent third-party COA and our supplier's COA on our Lab Results page.

What a COA actually is

A certificate of analysis is a lab report issued for a specific batch, or lot, of material. That word "lot" is the whole point. Kava is an agricultural product. Two harvests from the same farm, same cultivar, same age of root, will not test identically. Rainfall changes things. Drying changes things. Which portion of the root system went into the grind changes things a lot.

So a COA is a snapshot of one production run. It generally covers three areas:

• Identity. Confirming the material is genuinely Piper methysticum, and specifically the root, rather than a substitute, a filler, or the wrong part of the right plant.
• Composition and potency. The six major kavalactones measured individually, a total kavalactone figure, and usually the flavokavains.
• Purity. Heavy metals, microbial counts, pesticide residue, residual solvents, mycotoxins, and moisture content.

Most of that testing is done by HPLC, high-performance liquid chromatography, which separates a sample into its individual compounds and measures each one. A validated HPLC-UV method can resolve the six kavalactones and three flavokavains in about ten minutes.^[1] This is not exotic science. It's routine analytical chemistry, and the fact that it's routine is exactly why a brand having no COA is a choice rather than a limitation.

Why kava earned its need for testing

Most botanicals get tested because it's good practice. Kava gets tested because of what happened when nobody was.

In 2002, following a cluster of case reports of liver injury, Germany's federal drug regulator pulled every kava product from the market. The UK followed. The FDA issued a consumer advisory and suspended clinical work.^[2] For a plant with roughly three thousand years of daily ceremonial use behind it and no comparable history of liver problems in the Pacific, that was a strange outcome, and researchers spent the next decade working out why.

The answer that emerged had less to do with kava and more to do with what was being sold as kava. Several European manufacturers had been using aerial parts of the plant, meaning leaves, stems, and stem peelings, rather than root. It's cheaper and it increases yield. It also changes the chemistry completely. The aerial parts carry almost no kavalactones, but they do carry an alkaloid called pipermethystine, which concentrates in exactly those tissues, and they carry flavokavains at levels far above what's in the root.^[3]

There were other contributing factors, including solvent extraction methods that concentrate compounds traditional water preparation leaves behind, and at least some cases involving heavy concurrent alcohol use or existing medication load. But the raw material problem is the one that matters here, because it's the one a lab report catches before the product ships.

Germany's ban was eventually overturned in court, and kava was unscheduled there in 2014.^[4] The reputational damage outlasted the ban by a decade. Every kava company operating today is still doing business in the shadow of a problem that identity and composition testing would have prevented.

The one thing a label can't prove

Kava cultivars are broadly sorted into noble and tudei. Noble cultivars are the ones Pacific Island communities selected over centuries for daily drinking: balanced, pleasant, and clean the next morning. Tudei, from "two day," is faster growing, cheaper, higher in total kavalactones, and known for lingering heaviness and nausea. We wrote a full breakdown in noble vs. tudei kava.

Here's the problem. A bag of ground kava root does not announce which one it is. Ground tudei looks like ground noble. It's cheaper, and it's blended into noble stock often enough that the practice has a name in the industry.

What separates them is measurable. Tudei cultivars produce substantially more flavokavain B, the compound most implicated in the liver-toxicity concerns of the early 2000s. A working definition that's held up in peer review treats kava as non-noble when the ratio of flavokavains (A, B, and C) to kavalactones exceeds 1.^[1] Modern HPLC methods quantify flavokavains down to about 0.01 mg/g, which is sensitive enough to catch a noble batch that's been cut with tudei rather than only catching pure tudei.^[1]

There's a rough field version too. Drop a sample in acetone and noble material tends to turn the liquid yellow, while non-noble material pushes it orange, because of those higher red-brown flavokavains. It's a useful screen. It is not a substitute for a number on a report.

How to read a kava COA, line by line

Open a real one and it looks like a tax form. Here's what to actually look at, in order.

1. The header block

Before any chemistry, check four things: the name of the laboratory, the sample or lot identifier, the date the sample was received, and the date the report was issued. If the lot number on the report doesn't correspond to something you can find on the product, the rest of the document is decoration. This is the single most common gap, and it's the easiest one to check.

2. Identity

The report should confirm the material is Piper methysticum. Better reports specify root and rhizome. This is the line that would have flagged the aerial-parts substitution in 2002, and it's the reason identity testing exists as a separate category from potency.

3. Kavalactone potency

You'll see six compounds listed individually, plus a total. The six major kavalactones, with their conventional numbering:

• 1. Desmethoxyyangonin (DMY)
• 2. Dihydrokavain (DHK)
• 3. Yangonin (Y)
• 4. Kavain (K)
• 5. Dihydromethysticin (DHM)
• 6. Methysticin (M)

Those numbers aren't arbitrary rankings of importance. They're a fixed index, and they're what makes the next section readable. For more on what each one does, see our guide to what kavalactones actually are.

The total figure is the one most brands advertise. Treat it as a starting point rather than a grade. Higher total kavalactones is not automatically better, and it's worth knowing that tudei kava typically tests higher in total kavalactones than noble kava does. Potency alone can point the wrong direction.

4. The chemotype

Rank those six by abundance, most to least, and write the index numbers in order. That six-digit string is the chemotype, and it's the most useful single line on a kava COA.^[5]

The first two or three digits carry most of the meaning. A chemotype opening with 4 is kavain-dominant, which tends to read as clearer and more sociable. One opening with 2 is dihydrokavain-dominant and tends to sit heavier and more sedative. Chemotype tells you what an evening with that batch will actually feel like, which is more information than any total-kavalactone number gives you.

5. Flavokavains

Frequently the section brands skip. Flavokavains A, B, and C should be quantified, and you want the flavokavain-to-kavalactone ratio well under 1. This is the noble verification discussed above. A COA with a full kavalactone breakdown and no flavokavain panel is answering the easy question and leaving the important one open.

6. Heavy metals

Lead, arsenic, cadmium, mercury, at minimum. Kava is a root crop, and roots pull up what's in the soil, so this is not a box-ticking exercise. For reference on how tight the thresholds are, the NSF/ANSI 173 standard caps lead in a finished dietary supplement at 10 micrograms per daily dose.^[6] Published survey work on botanical supplements, kava included, has generally found metal levels within acceptable ranges, but "generally" is doing real work in that sentence and it's why per-batch testing exists.^[7]

7. Microbials

Total plate count, yeast and mold, coliforms, and specifically E. coli, Salmonella, and Staphylococcus aureus. Kava is dried and shipped across an ocean, sometimes stored in humid conditions on both ends. That same survey work on botanical supplements found bacterial or fungal presence in most samples analyzed, which is normal for dried plant material and exactly why limits and testing exist.^[7]

8. Pesticides, solvents, mycotoxins, moisture

Pesticide screening matters more for imported agricultural material than people assume. Residual solvents only apply if extraction was involved, and if a product is extracted, you want that panel. Mycotoxins, especially aflatoxins, track with storage conditions. Moisture content sounds like the boring line item and isn't: high moisture in stored kava is what invites mold in the first place.

"Third-party" is doing a lot of work in that sentence

Not every COA is independent, and the difference is worth understanding rather than assuming.

A supplier COA comes from the vendor who sold the raw material, or from a lab they hired. It's genuinely useful. It tells you the supplier is running quality control and it documents the material as it arrived. But the party issuing it has a commercial interest in the result being good.

A third-party COA means the brand selling the finished product paid an independent lab, with no stake in the outcome, to test the batch that actually goes in the bottle. Those are different documents answering different questions, and the strongest position is publishing both, because it lets you see whether what the supplier claimed survived into the finished product.

Three things separate a real independent report from a document any brand could generate:

• ISO/IEC 17025 accreditation. This is the credential that means a lab's methods and competence have been audited by an outside accreditation body. It's the clearest single marker of a legitimate testing facility.^[8]
• Lot-specific reporting. One COA presented as representative of an entire product line is closer to a marketing asset than a quality record.^[8]
• A document you can open. Not a badge graphic. Not a sentence on the FAQ page. The actual PDF, with the lab's name on it.

It's worth knowing that "dry labbing" is a documented practice in supplement testing, where a lab issues reports without running the analyses.^[9] That's an argument for accredited labs and for brands publishing full reports rather than summary numbers, since a full report is a much harder thing to fake convincingly.

Red flags on a kava COA

• "Third-party tested" with nothing published. The phrase is unregulated. If there's no document, treat the claim as absent.
• No lab name on the report. An unattributable test result is not a test result.
• No lot number, or a lot number that matches nothing on the product.
• One COA covering everything. Different SKUs and different production runs need different reports.
• An old report on a current product. Check the issue date against what you're buying.
• Potency only, no purity panel. Kavalactone numbers are the fun part. Heavy metals and microbials are the safety part.
• No flavokavain data on a product marketed as noble. The nobility claim rests on that panel.
• A cropped screenshot instead of the full document. Ask what got cropped out.

What we publish, and why

I'd rather show than assert, so here's ours.

We publish two documents on our Lab Results page: our own independent third-party COA and our supplier's COA, side by side. Both open as full PDFs. You can read them without contacting us or joining anything.

The kavalactone profile on our third-party report comes back at 294mg of total kavalactones per 44 mL shot, with an abundance ranking that leads kavain, then dihydrokavain, then yangonin. A chemotype opening 4-2-3 is a kavain-dominant profile, which is the signature of noble kava chemistry and lines up with the clearer, more balanced experience noble cultivars are selected for.

We also publish the residual solvent panel, which is the part most brands would quietly leave out. Ethanol and methanol were the only compounds detected above the lab's reporting limit. Rather than hide that, we did the math on it: each shot contains 0.0149g of ethanol, so you'd need around 940 of them to equal one standard alcoholic drink, and 2.76mg of methanol, which puts the EPA's long-term daily reference threshold at roughly 58 shots a day. Everything else came back below reporting limits.

Those numbers are not impressive marketing. They're just true, and publishing the boring true version is the entire point of a COA. A lab report you only share when every line is flattering isn't transparency, it's a press release.

The bottom line

Kava is a plant with a genuinely good safety record when it's the right part of the right cultivar, prepared the right way. The moments it has gone wrong in the West trace back to the supply chain rather than to kava itself, and a certificate of analysis is the specific instrument that makes a supply chain legible.

The industry has largely converged on saying "third-party lab tested" and largely not converged on publishing the tests. That gap is where you should be doing your shopping. Ask for the document. Check that it matches your lot. Read past the total kavalactone number to the chemotype and the flavokavains. It takes about ninety seconds once you know where to look, and it's the difference between trusting a brand and verifying one.

Frequently Asked Questions

Sources

1. Wang Y, et al. Single-Laboratory Validation for the Determination of Kavalactones and Flavokavains in Piper methysticum (Kava) (2018), Journal of AOAC International. PubMed
2. National Institute of Diabetes and Digestive and Kidney Diseases. Kava Kava, LiverTox: Clinical and Research Information on Drug-Induced Liver Injury. NCBI Bookshelf
3. Rowe A, Zhang LY, Ramzan I. Toxicokinetics of Kava (2011), Advances in Pharmacological Sciences. Wiley Online Library
4. Teschke R, Schulze J, Schwarzenboeck A, et al. German Kava Ban Lifted by Court: The Alleged Hepatotoxicity of Kava (Piper methysticum) as a Case of Ill-Defined Herbal Drug Identity, Lacking Quality Control, and Misguided Regulatory Politics (2015). PubMed
5. Tang Y, et al. Kavalactone content and chemotype of kava beverages prepared from roots and rhizomes of Isa and Mahakea varieties (2015). PMC
6. NSF International. NSF/ANSI 173: Dietary Supplements (heavy metal limits for finished dietary supplement products). NSF.org
7. Dolan SP, Nortrup DA, Bolger PM, Capar SG. Evaluation of Metal and Microbial Contamination in Botanical Supplements (2003), Journal of Agricultural and Food Chemistry. PubMed
8. International Organization for Standardization. ISO/IEC 17025: General requirements for the competence of testing and calibration laboratories. ISO.org
9. ConsumerLab.com. Can I trust supplement manufacturer lab reports and certificates of analysis? ConsumerLab
10. US Food and Drug Administration. Scientific Memorandum: Kava (2020). FDA.gov