What 'Standardized' Actually Means on a Botanical Extract COA — and When the Numbers Lie

A turmeric extract that says “95% curcuminoids” on the COA isn’t necessarily lying. It’s also not necessarily telling you what you think it is.

This is the quiet problem in botanical raw material sourcing: standardization language has drifted so far into marketing territory that the number on the certificate almost requires a second certificate to interpret it. After reviewing hundreds of vendor COAs for botanical extracts, the pattern is consistent — the specification is listed, the test method is referenced, and the pass/fail determination is right there in the results column. But ask three different suppliers what “standardized” means and you’ll get three different answers, because there genuinely isn’t one. The term isn’t defined in DSHEA, it doesn’t carry a USP-mandated definition for all dietary botanical ingredients, and AHPA’s guidance, while useful, isn’t enforceable.

So the number on your COA reflects whatever your vendor decided standardized means on the day they wrote their test method.

The Three Things “Standardized” Can Actually Mean

In botanical manufacturing, the word “standardized” can describe at least three distinct situations — and confusing them leads to real formulation and compliance problems.

Marker-standardized extracts are adjusted during processing to hit a defined concentration of one or more chemical markers. The marker may or may not be the bioactive driving the clinical effect. Milk thistle standardized to 70–80% silymarin is a well-grounded example: silymarin (a complex of flavonolignans) is considered the primary hepatoprotective constituent. Valerian standardized to 0.3–0.8% valerenic acid is more ambiguous — valerenic acid is a useful identity and potency marker, but the herb’s sleep-related activity is almost certainly multifactorial and valerenic acid alone doesn’t tell the full story.

Full-spectrum extracts with marker verification retain the plant’s complete chemical profile through gentle extraction while using marker compound testing to confirm identity and approximate potency. Ashwagandha root is often produced this way, with 5% withanolides cited as a minimum threshold. Trademarked ingredients like KSM-66 and Sensoril back their full-spectrum claims with proprietary clinical and analytical data — which is partly why their price reflects a premium over unbranded “5% withanolides” material of uncertain origin.

Ratio extracts — expressed as 10:1, 20:1, or similar — indicate how much raw herb was concentrated into the final extract by weight. A 10:1 ratio means 10 kg of raw herb was processed down to 1 kg of extract. Here’s the problem: ratio alone tells you nothing about potency. Two 10:1 extracts from two different suppliers, run through the same analytical testing laboratory side by side, can show dramatically different marker compound concentrations depending on the starting material quality, harvest conditions, and extraction method. A 10:1 ratio from high-quality, well-timed ashwagandha root can outperform a 20:1 from low-grade material. The ratio is a process descriptor, not a potency guarantee.

What Third-Party Verification Actually Looks Like

When our team receives a botanical extract for incoming raw material testing, the vendor’s COA is a starting point — not a conclusion. Here’s what rigorous third-party verification involves and why the choices matter.

HPLC for marker compound quantification. High-performance liquid chromatography with UV or mass spectrometric detection is the reference-grade method for most botanical markers. For curcuminoids in turmeric, AOAC Method 2012.22 and USP’s turmeric extract monograph define the framework. Critically, the specific HPLC conditions matter: curcuminoid analysis at 420 nm versus 425 nm, the reference standard grade used for calibration, and whether the method resolves the three major curcuminoids individually (curcumin, demethoxycurcumin, bisdemethoxycurcumin) all affect the result. A method that reports “total curcuminoids” as a single summed number without resolving individual peaks is less informative and more susceptible to interference.

UV spectrophotometry — faster, cheaper, and often misleading. Some vendor COAs report marker concentrations by UV spectrophotometry, which is quicker and requires less sophisticated instrumentation. For straightforward, isolated compounds in a clean matrix, UV spec can be adequate. But for complex botanical extracts containing groups of structurally related compounds — like curcuminoids or silymarin — UV spectrophotometry systematically overcounts because it captures UV-absorbing compounds beyond the target analytes. We consistently see differences of 8–12 percentage points between UV-spec and HPLC results on the same extract. A material reported at 95% curcuminoids by UV spec may come in at 83–87% under a validated HPLC method. That’s not fraud on the supplier’s part — it’s a method selection issue. But the result lands in your specifications and your label, not theirs.

HPTLC and DNA barcoding for identity. Even when marker compound quantification passes, a separate identity test is essential. An adulterant that happens to share UV-absorbing chemistry won’t be caught by potency testing alone. High-performance thin-layer chromatography (HPTLC) generates a chemical fingerprint of the extract’s profile compared against authenticated reference standards from sources like the American Herbal Pharmacopoeia (AHP) or USP’s Reference Standards program. DNA barcoding provides species-level identity confirmation — especially useful for detecting substitution with morphologically similar botanicals. Both methods provide orthogonal evidence that the certificate’s identity claim is accurate. Neither is redundant.

A Midwest Brand’s $40,000 Lesson

The sourcing pressure is real, and the margins in supplement manufacturing are thin enough that a 12% cost advantage on a high-volume botanical ingredient is genuinely significant. We consulted with a Chicago-area brand that sourced an ashwagandha root extract specified at 5% withanolides from a new supplier offering that exact discount over their existing vendor. The COA looked clean: 5.2% withanolides, test method listed as “UV spectrophotometry,” passing results across the standard panel, dated and signed.

Third-party ICP-MS testing flagged elevated lead at 1.8 ppm. At the brand’s intended 600 mg daily serving, that translates to approximately 1.08 µg lead per dose — well below USP <232>‘s 5 µg/day limit for lead. But their finished product included two additional botanical ingredients also sourced from new suppliers, and cumulative lead exposure across the full formula was the actual compliance risk. Independent HPLC marker testing came in at 3.9% withanolides — not 5.2%. And HPTLC identity analysis confirmed ashwagandha as the primary species but flagged trace chromatographic signatures consistent with a filler botanical mixed into the lot.

The 12% price advantage evaporated. Reformulation, batch hold, and the delayed launch timeline cost the brand considerably more than the per-kilo savings would have provided over two years of purchasing.

This scenario is more common than the industry publicly acknowledges. A peer-reviewed analysis of commercial botanical dietary ingredients in the US market estimated that approximately 34% of tested extracts failed to meet their stated potency specifications when re-analyzed by an independent laboratory. That figure is consistent with what commercial analytical testing laboratories see in day-to-day incoming raw material work, and it explains why supplier qualification programs — not just supplier approval programs — exist in well-run supplement operations.

Five Things to Check on a COA Before You Send Samples to the Lab

A pre-screening review of vendor documentation can flag obvious problems before you spend testing budget confirming them.

1. Method specificity. Does the COA name the exact analytical method? “HPLC” isn’t enough — look for the reference framework (AOAC, USP, in-house validated), the detection mode (UV, MS, ELSD), and where possible, the reference standard lot used for calibration. “In-house method” without a validation reference should prompt a method transfer request.

2. Single-marker vs. profile reporting. A turmeric COA reporting only “total curcuminoids” is less informative than one resolving the three curcuminoids individually. The curcumin:demethoxycurcumin:bisdemethoxycurcumin ratio tells you something about the extract’s provenance and processing that a single number doesn’t.

3. Specification type. Is the specification “NLT 95%” or “95% ± 2%”? A minimum-only spec is standard in botanical ingredient trade, but for formulation purposes you need to understand the range your batch might actually fall in — which requires historical lot data from the supplier.

4. Per-lot vs. template COAs. Template COAs carry fixed values regardless of the actual lot tested — a clear indicator that per-lot testing either isn’t happening or isn’t being documented traceably. Lot-specific documents should carry a unique batch/lot number that traces directly to manufacturing and testing records.

5. Third-party accreditation status. A COA issued by the manufacturer’s own quality lab and a COA from an ISO 17025–accredited analytical testing laboratory represent meaningfully different levels of assurance. For high-risk raw materials — Ayurvedic herbs, extracts from markets with limited agricultural traceability, or any botanical with known adulteration history — ISO 17025–accredited third-party data is the appropriate reference document.

A Practical Incoming Testing Protocol for Botanical Extracts

For brands sourcing standardized botanical extracts — particularly those with potent marker specifications or Ayurvedic or Chinese herbal origin — the minimum defensible testing protocol looks like this:

• New supplier qualification: HPTLC + DNA barcoding identity, HPLC marker quantification, full USP <232>/<233> elemental impurity panel (ICP-MS), and USP <61>/<62> microbial limits on each of the first three lots from that supplier
• Established supplier, ongoing: Skip-lot HPLC marker quantification after five consecutive passing lots; retain per-lot microbial and heavy metals testing for high-risk botanicals; per-lot identity testing where adulteration risk is elevated (e.g., turmeric, saffron, elderberry)
• Documentation for DSHEA audit readiness: Test results linked to specific lot numbers, certificates of analysis from the testing laboratory in the file, written specification with the acceptable range, and out-of-specification (OOS) investigation records where applicable

This is the framework an FDA dietary supplement inspection expects to find documented under 21 CFR Part 111. It’s also what protects your brand when a contamination or potency complaint works its way upstream and your supplier’s response is a form letter pointing back to their own COA.

The number on the certificate is the starting point. What it means — and whether it’s accurate — is what third-party testing is for.

---

Written by Nour Abochama, VP Operations, Qalitex | Quality Consultant, Ayah Labs. Learn more about our team

Ship your sample to our Chicago facility — get a Qalitex CoA in 5–7 days. Contact us

Related from our network

• ISO 17025–Accredited Supplement Testing at Qalitex Laboratories — Full-service raw material and finished product testing under ILAC-recognized accreditation, with sample intake in Los Angeles and results delivered digitally.