🔬 How to read a kratom COA: a vendor's perspective 🧪
Posted: Tue May 05, 2026 12:12 pm
Hey all. Been a minute since I posted on here, I just pop in once in a while to see what's going on. Been seeing a lot of confusion in various threads lately about how to actually read a kratom COA, what the different standards mean, and what separates real QC from window dressing. Figured I'd write up a primer since I've been running Hudson Valley Botanicals (Garuda Kratom) since 2017 and this stuff is what I deal with on the operations end all the time. The recent DTE thread is what prompted this since a lot of the technical questions came up there, but I want to keep this post broader than just one vendor.
Not trying to dunk on a competitor, just figured the educational context might be useful for people trying to evaluate any kratom vendor's testing.
On reading a COA
Different labs format reports differently and not every COA shows action limits or pass/fail status, which is fine as long as you can verify the methodology. What you can always look for: the lab's accreditation (ISO 17025 is the standard you want), the specific testing methods cited (AOAC reference numbers for microbial, ICP-MS for heavy metals), the LOD and LOQ values so you understand what the instrument can actually detect, and a signature from the analyst and lab director. Then you compare the actual numbers to whatever standard applies to the product category.
A practical tip: real transparency means COAs are available without you having to ask. Public dashboards, downloadable PDFs per batch, lot numbers you can cross-reference. If a vendor won't share lab results, only shows one COA for the whole product line, or hand-waves with "we test, trust us," that's a red flag. There's no good reason in 2026 to not publish per-batch COAs.
On heavy metals
There's no FDA-specific limit for kratom because kratom isn't formally regulated as a dietary supplement. The framework most of the industry references is USP <2232>, which sets daily oral exposure limits of roughly 5 mcg/day for lead, 15 for inorganic arsenic, 5 for cadmium, and 30 for mercury. You back-calculate from serving size to get a ppm limit on the powder. ICP-MS is the gold standard analytical method, parts-per-billion sensitivity. California Prop 65 is way stricter on lead at 0.5 mcg/day, which is why some vendors won't ship there.
For example, our last four batches tested via ICP-MS at Harken Research came in with lead between 0.391 and 0.745 µg/g, cadmium below the limit of quantitation across the board, mercury under 0.04 µg/g. To put the lead numbers in context: at 6 grams (the "do not exceed daily" amount on our label), our highest reading delivers under 4.5 mcg of lead, which is below the USP <2232> daily exposure limit of 5 mcg. WHO's guideline is 10 mg/kg (10 µg/g), so we're running 13-25x under that. All results are public on the dashboard at https://hudsonvalleybotanicals.com/coas/ if anyone wants to see the actual reports.
On microbial limits
This is where it gets technical because there's a real tier debate in the industry. AHPA's botanical microbiology guidance has different limits depending on how the product is categorized. Raw, unprocessed dried herb gets looser limits. Finished, ready-to-consume product gets stricter ones. Most kratom vendors treat plain leaf powder as the raw-herb category since it's a single-ingredient unprocessed botanical, even though it's sold ready to consume. There's reasonable disagreement here, but our published release specifications align with AHPA's finished-product tier rather than the looser raw-herb tier, since our product is sold in sealed pouches with no further processing by the customer.
What there's no disagreement on: salmonella must be absent. E. coli must be absent. Both AHPA tiers, USP, FDA guidance under 21 CFR Part 111, the AKA GMP standard, all of them. Pathogen presence is the line that doesn't move based on category. If a batch comes back salmonella detected, the protocol is destroy or remediate (irradiation, typically) and re-test. You don't release it. That part isn't a tier debate, it's the floor.
For reference, our Total Plate Counts on the most recent four batches ran between <100 and 800 CFU/g. Coliforms <100 across the board. Salmonella and E. coli negative on all of them. To put those in context: the finished-product tier limits are 10,000 CFU/g for aerobic count and 100 CFU/g for coliforms, so we're running at least an order of magnitude under both. Low numbers here generally reflect supplier hygiene at the source, since you can't really fix high microbial counts after the fact without remediation steps that affect alkaloid integrity.
On what usually happens when a batch tests positive
For context, since this comes up rarely enough that people aren't always sure what's standard practice. Salmonella detection in kratom does happen occasionally because it's a botanical grown and processed in tropical environments, but the industry response when it's caught is well-established. The 2017-2018 multi-state outbreak resulted in 13 recalls covering 67 products from 26 distributors. Most were voluntary, where the vendor pulled product and worked with FDA. One vendor (Triangle Pharmanaturals) refused to cooperate and got hit with the first-ever mandatory FDA food recall in U.S. history, and went out of business shortly after. As recently as October 2025, Bedrock MFG voluntarily recalled their Monarch Premium Kratom line nationwide after the Florida Department of Agriculture flagged salmonella in a sample. They worked with FDA and pulled product from retailers.
The point being: there's a well-worn playbook when this comes up. Voluntary recall, public notice, work with regulators, replace product. Eighteen states now have Kratom Consumer Protection Acts that explicitly require batch microbial testing and immediate recall for non-compliant product. So when you see a salmonella detection on a published COA, the question isn't whether it's a problem (it is, under any standard), the question is what the vendor did about it. Searching FDA's recall database is one way to check.
On QC in practice, since this is about more than just standards
Three independent panels per batch: microbial, heavy metals via ICP-MS, and alkaloid profile by HPLC. We use ISO 17025 accredited labs. If a batch fails any panel, it doesn't ship. We post the actual reports.
The supplier side is the part nobody really talks about because it's not exciting content. Last few months I had to dump 50kg-100kg of a batch because there was way too much fibrous material in the grind. Likely stem and vein getting through their filtration setup, which to me means their grinding and screening process isn't tight enough. Alkaloid content might've been fine, but if that much fiber is slipping through, what else is? So it got tossed. Four figure loss on a single batch. I am no longer sourcing from that supplier.
I cycle suppliers constantly. Quality slips, I move on. What I'm looking for: GMP compliant facility, years in business, references from other importers I trust, willingness to do third-party verification, consistency batch to batch, and lab results I can independently confirm. Indonesian export regs have genuinely tightened up post-regulation, that part's true. But passing export testing is the floor, not the ceiling. The LS document confirms a batch passed export inspection. It tells you nothing about what happened to the material before it got crated up or after it landed stateside.
On the enhanced leaf question
Since this comes up periodically and was raised again recently, quick technical answer. You wouldn't necessarily see crystals in spiked powder. Extracts can be dissolved in solvent, sprayed on leaf, and dried. Done halfway competently, nothing visible. Where it shows up is the alkaloid profile on the COA. Plain leaf has a predictable mitragynine to 7-OH ratio, with 7-OH usually well under 0.1% of total alkaloids. Typical mitragynine ranges from about 0.5% to 2% of dry weight. If a COA shows 7-OH elevated relative to mitragynine, or total alkaloid content implausibly high for raw leaf, that's the tell. Consistency of effect alone doesn't prove anything, good sourcing produces consistent leaf, but the alkaloid panel will tell you if something's off.
Anyway, didn't mean to write an essay. Hope some of this is useful context for evaluating kratom vendors generally. Take care out there.
Scott @ Hudson Valley Botanicals
Not trying to dunk on a competitor, just figured the educational context might be useful for people trying to evaluate any kratom vendor's testing.
On reading a COA
Different labs format reports differently and not every COA shows action limits or pass/fail status, which is fine as long as you can verify the methodology. What you can always look for: the lab's accreditation (ISO 17025 is the standard you want), the specific testing methods cited (AOAC reference numbers for microbial, ICP-MS for heavy metals), the LOD and LOQ values so you understand what the instrument can actually detect, and a signature from the analyst and lab director. Then you compare the actual numbers to whatever standard applies to the product category.
A practical tip: real transparency means COAs are available without you having to ask. Public dashboards, downloadable PDFs per batch, lot numbers you can cross-reference. If a vendor won't share lab results, only shows one COA for the whole product line, or hand-waves with "we test, trust us," that's a red flag. There's no good reason in 2026 to not publish per-batch COAs.
On heavy metals
There's no FDA-specific limit for kratom because kratom isn't formally regulated as a dietary supplement. The framework most of the industry references is USP <2232>, which sets daily oral exposure limits of roughly 5 mcg/day for lead, 15 for inorganic arsenic, 5 for cadmium, and 30 for mercury. You back-calculate from serving size to get a ppm limit on the powder. ICP-MS is the gold standard analytical method, parts-per-billion sensitivity. California Prop 65 is way stricter on lead at 0.5 mcg/day, which is why some vendors won't ship there.
For example, our last four batches tested via ICP-MS at Harken Research came in with lead between 0.391 and 0.745 µg/g, cadmium below the limit of quantitation across the board, mercury under 0.04 µg/g. To put the lead numbers in context: at 6 grams (the "do not exceed daily" amount on our label), our highest reading delivers under 4.5 mcg of lead, which is below the USP <2232> daily exposure limit of 5 mcg. WHO's guideline is 10 mg/kg (10 µg/g), so we're running 13-25x under that. All results are public on the dashboard at https://hudsonvalleybotanicals.com/coas/ if anyone wants to see the actual reports.
On microbial limits
This is where it gets technical because there's a real tier debate in the industry. AHPA's botanical microbiology guidance has different limits depending on how the product is categorized. Raw, unprocessed dried herb gets looser limits. Finished, ready-to-consume product gets stricter ones. Most kratom vendors treat plain leaf powder as the raw-herb category since it's a single-ingredient unprocessed botanical, even though it's sold ready to consume. There's reasonable disagreement here, but our published release specifications align with AHPA's finished-product tier rather than the looser raw-herb tier, since our product is sold in sealed pouches with no further processing by the customer.
What there's no disagreement on: salmonella must be absent. E. coli must be absent. Both AHPA tiers, USP, FDA guidance under 21 CFR Part 111, the AKA GMP standard, all of them. Pathogen presence is the line that doesn't move based on category. If a batch comes back salmonella detected, the protocol is destroy or remediate (irradiation, typically) and re-test. You don't release it. That part isn't a tier debate, it's the floor.
For reference, our Total Plate Counts on the most recent four batches ran between <100 and 800 CFU/g. Coliforms <100 across the board. Salmonella and E. coli negative on all of them. To put those in context: the finished-product tier limits are 10,000 CFU/g for aerobic count and 100 CFU/g for coliforms, so we're running at least an order of magnitude under both. Low numbers here generally reflect supplier hygiene at the source, since you can't really fix high microbial counts after the fact without remediation steps that affect alkaloid integrity.
On what usually happens when a batch tests positive
For context, since this comes up rarely enough that people aren't always sure what's standard practice. Salmonella detection in kratom does happen occasionally because it's a botanical grown and processed in tropical environments, but the industry response when it's caught is well-established. The 2017-2018 multi-state outbreak resulted in 13 recalls covering 67 products from 26 distributors. Most were voluntary, where the vendor pulled product and worked with FDA. One vendor (Triangle Pharmanaturals) refused to cooperate and got hit with the first-ever mandatory FDA food recall in U.S. history, and went out of business shortly after. As recently as October 2025, Bedrock MFG voluntarily recalled their Monarch Premium Kratom line nationwide after the Florida Department of Agriculture flagged salmonella in a sample. They worked with FDA and pulled product from retailers.
The point being: there's a well-worn playbook when this comes up. Voluntary recall, public notice, work with regulators, replace product. Eighteen states now have Kratom Consumer Protection Acts that explicitly require batch microbial testing and immediate recall for non-compliant product. So when you see a salmonella detection on a published COA, the question isn't whether it's a problem (it is, under any standard), the question is what the vendor did about it. Searching FDA's recall database is one way to check.
On QC in practice, since this is about more than just standards
Three independent panels per batch: microbial, heavy metals via ICP-MS, and alkaloid profile by HPLC. We use ISO 17025 accredited labs. If a batch fails any panel, it doesn't ship. We post the actual reports.
The supplier side is the part nobody really talks about because it's not exciting content. Last few months I had to dump 50kg-100kg of a batch because there was way too much fibrous material in the grind. Likely stem and vein getting through their filtration setup, which to me means their grinding and screening process isn't tight enough. Alkaloid content might've been fine, but if that much fiber is slipping through, what else is? So it got tossed. Four figure loss on a single batch. I am no longer sourcing from that supplier.
I cycle suppliers constantly. Quality slips, I move on. What I'm looking for: GMP compliant facility, years in business, references from other importers I trust, willingness to do third-party verification, consistency batch to batch, and lab results I can independently confirm. Indonesian export regs have genuinely tightened up post-regulation, that part's true. But passing export testing is the floor, not the ceiling. The LS document confirms a batch passed export inspection. It tells you nothing about what happened to the material before it got crated up or after it landed stateside.
On the enhanced leaf question
Since this comes up periodically and was raised again recently, quick technical answer. You wouldn't necessarily see crystals in spiked powder. Extracts can be dissolved in solvent, sprayed on leaf, and dried. Done halfway competently, nothing visible. Where it shows up is the alkaloid profile on the COA. Plain leaf has a predictable mitragynine to 7-OH ratio, with 7-OH usually well under 0.1% of total alkaloids. Typical mitragynine ranges from about 0.5% to 2% of dry weight. If a COA shows 7-OH elevated relative to mitragynine, or total alkaloid content implausibly high for raw leaf, that's the tell. Consistency of effect alone doesn't prove anything, good sourcing produces consistent leaf, but the alkaloid panel will tell you if something's off.
Anyway, didn't mean to write an essay. Hope some of this is useful context for evaluating kratom vendors generally. Take care out there.
Scott @ Hudson Valley Botanicals
. Lately I’ve been wondering about lead and the potential heavy metals I may have been consuming for years.