Botanical Testing: What Supplement Brands Should Request to Verify Ingredient Quality and Screen for Contaminants

Supplement brands should request four categories of botanical testing on every herbal ingredient: identity confirmation, strength or marker assay, contaminant screening, and targeted adulterant screening. Identity testing is required by law for every incoming lot of a dietary ingredient under 21 CFR 111.75. Contaminant screening should cover heavy metals, pesticide residues, microbial limits, and, where the material warrants it, mycotoxins, residual solvents, and other process-related contaminants.

The right panel depends on the form of the material. A whole dried root, a powdered herb, and a solvent-extracted standardized extract carry different risks and call for different methods. A brand that specifies these tests in its purchase specifications, and confirms them through a qualified laboratory, meets the core expectations of the FDA dietary supplement GMP rule and retailer programs such as Amazon's.

What Is Botanical Testing?

Botanical testing is the laboratory analysis of plant-derived ingredients used in dietary supplements, foods, and natural health products. It confirms the identity of the plant species and part, measures the strength of marker compounds, and screens for contaminants such as heavy metals, pesticides, microorganisms, mycotoxins, and residual solvents. It also detects undeclared adulterants.

The Core Botanical Testing Panel at a Glance

Frequency and depth vary by material type and risk, which later sections address.

• Identity: Confirms the correct species and plant part. Methods include macroscopic and microscopic examination, HPTLC, HPLC fingerprinting, FTIR or NIR spectroscopy, and DNA methods for unprocessed material.
• Strength (assay): Quantifies marker or active compounds, usually by HPLC-UV or LC-MS/MS, against a numeric specification.
• Heavy metals: Arsenic, cadmium, lead, and mercury by ICP-MS, evaluated against the daily dose.
• Pesticide residues: Multi-residue screen by LC-MS/MS and GC-MS/MS, often several hundred compounds per run.
• Microbial limits: Total aerobic count, yeast and mold, and absence of specified pathogens such as Salmonella and E. coli.
• Mycotoxins: Aflatoxins and ochratoxin A by LC-MS/MS for materials with storage or climate risk.
• Residual solvents: Gas chromatography per USP General Chapter <467> for solvent-extracted materials.
• Targeted adulterant screens: Undeclared drugs, synthetic stimulants, dyes, or cheaper plant substitutes, based on the known risk profile of the ingredient.

Why Identity Testing Comes First

Identity testing is the one botanical test that United States regulation requires on every lot. 21 CFR 111.75(a)(1) states that a manufacturer must conduct at least one appropriate test or examination to verify the identity of any dietary ingredient before use. A supplier certificate of analysis does not satisfy this requirement. The only exception is a petition to FDA for an alternative testing plan, which must show no material loss of assurance compared with 100 percent testing.

Compendial guidance favors orthogonal identity testing, meaning two or more methods that rely on different principles. A review of compendial perspectives on botanical identity testing notes that USP chapters combine macroscopic, microscopic, and chemical fingerprint methods rather than relying on any single technique. Microscopy cannot identify a solvent extract, and DNA methods can fail on processed material.

Matching identity methods to the material

• Whole or cut plant material: Macroscopic and microscopic examination against a reference specimen remains a valid and inexpensive starting point. HPTLC adds a chemical fingerprint.
• Powdered herb: Microscopy still works when cell structures survive milling. HPTLC or HPLC fingerprinting confirms the chemical profile.
• Extracts: Anatomical features are gone, so chemical methods carry the weight. HPTLC (USP <203>) and HPLC against a botanical reference standard are the usual tools.
• DNA methods: DNA barcoding can confirm species in raw material, but DNA degrades during extraction. The same review indicates that DNA methods can produce false negatives for concentrated extracts and should not be used alone for them.

For a plain-language summary of the lot-by-lot requirement and how it connects to identity, purity, strength, and composition specifications, see the BSCG explainer on dietary ingredient identity testing under 21 CFR 111.

Strength and Marker Compound Assays

A strength assay measures how much of a defined compound or compound class is present. For a standardized extract, the label may state a percentage of curcuminoids, ginsenosides, or another marker. That claim must trace back to a validated quantitative method, typically HPLC-UV or LC-MS/MS, with a stated acceptance range.

HPTLC fingerprints are semi-quantitative. They can show that marker classes are present in roughly the expected proportion, but they do not replace a validated assay when a numeric label claim is made. Specifications should name the marker, the method, the units, and the acceptance range. Published work on plant-to-extract ratios and botanical extract labeling explains why the plant part, extraction solvent, and ratio also belong on the specification sheet.

Contaminant Testing: What to Request and Why

21 CFR 111.70 requires manufacturers to set limits on contaminants that could adulterate the finished product. The regulation does not fix numeric thresholds. Brands set them using pharmacopeial chapters, trade association guidance, retailer requirements, and destination-market regulations.

Heavy metals

Request arsenic, cadmium, lead, and mercury by ICP-MS on every botanical ingredient. Heavy metal risk varies by source and material type, as BSCG's guidance on shilajit, a mineral-rich natural material, illustrates. USP General Chapter <2232> Elemental Contaminants in Dietary Supplements sets permitted daily exposure values for these four elements, derived from FAO/WHO tolerable intake figures. USP notes that the limits apply to finished dosage forms and that other rules, such as California Proposition 65, may impose different limits.

Because limits are expressed per day, the same ingredient result can pass in one product and fail in another with a higher serving size. Ask the laboratory to report results in µg/g with a stated limit of quantitation, then calculate exposure at the labeled daily dose. BSCG's testing services page describes its standard metals panel as arsenic, cadmium, chromium, lead, and mercury by ICP-MS, applying both USP/ICH and NSF/ANSI 173 thresholds.

Pesticide residues

Request a multi-residue pesticide screen for any agricultural botanical, and ask which compounds and which regulatory list the laboratory used. USP General Chapter <561> Articles of Botanical Origin provides pesticide limits and refers to domestic tolerances. In the European Union, Regulation (EC) 396/2005 applies a default maximum residue level of 0.01 mg/kg where no specific level has been set. A laboratory method must reach at least that level to demonstrate compliance for EU-bound products.

Retailer programs add their own expectations. BSCG's summary of the Amazon dietary supplement policy reports that botanical-based products require pesticide screening, including glyphosate, alongside heavy metals and microbial testing, using NSF/ANSI 173 or USP as the reference framework.

Microbial limits

Request total aerobic microbial count, total combined yeasts and molds, and absence tests for Salmonella, Escherichia coli, and Staphylococcus aureus. USP chapters <2021>, <2022>, and <2023> define the test procedures and acceptance criteria for dietary supplements. Dried plant material carries a higher fungal load than most other supplement ingredients, so yeast and mold results deserve close review. BSCG's microbial limits guide covers how to set raw material, in-process, and finished product microbial testing based on risk.

Mycotoxins

Request aflatoxin (B1, B2, G1, G2) and ochratoxin A testing if the commodity, source, or storage history indicates mycotoxin risk. The FDA action level for total aflatoxins in food is 20 ppb. EU limits for dried herbs and spices are lower. Validated LC-MS/MS methods can quantify aflatoxins at or below 1 to 2 ppb, so the method should be capable of reporting well under the applicable limit.

Residual solvents

Request residual solvent testing by gas chromatography for any extract made with organic solvents. USP General Chapter <467> groups solvents into three classes. Class 1 solvents such as benzene should be avoided, Class 2 solvents carry defined limits, and Class 3 solvents carry higher limits. Ask the supplier which solvents were used in extraction so the laboratory can target them.

Other process and source contaminants

Certain botanicals warrant additional screens. Polycyclic aromatic hydrocarbons (PAHs) are a documented contaminant in marketed supplements; a published survey of 94 food supplements found EU marker PAHs above the limit of quantification in many of the samples tested. Pyrrolizidine alkaloids are regulated in the European Union, where Regulation (EU) 2023/915 sets maximum levels for these alkaloids in herbal food supplements. Foreign matter inspection, moisture, and ash also appear in pharmacopeial botanical monographs as basic quality checks.

Adulterant Screening: Beyond Standard Contaminants

Standard contaminant panels do not detect deliberate adulteration. Adulteration takes three main forms in botanicals: substitution with a cheaper plant, spiking with a purified compound to pass a marker assay, and addition of undeclared synthetic drugs or stimulants.

The ABC-AHP-NCNPR Botanical Adulterants Prevention Program (BAPP) publishes ingredient-specific bulletins and laboratory guidance documents. Its bulletins cover ingredients including ashwagandha, bilberry, black cohosh, elderberry, ginkgo, ginseng, saw palmetto, and turmeric. Brands sourcing any of these should review the relevant bulletin and request the analytical approach it recommends, since standard identity tests may not distinguish a spiked or substituted material.

Synthetic stimulants have repeatedly appeared on labels under plant names. BSCG's Dietary Supplement Ingredient Advisory List documents examples such as BMPEA presented as Acacia rigidula extract, DMAA presented as geranium extract, and DMHA associated with purported Juglans regia or Kigelia africana extracts. Where an ingredient's category or history indicates risk, brands should request a targeted LC-MS/MS screen for undeclared drugs and stimulants. FDA's medication health fraud resources describe hidden prescription drug ingredients as an ongoing concern in products sold as supplements.

Testing Panels by Material Type

The following summary matches the panel to the material. It reflects compendial guidance and common industry practice rather than a fixed regulatory requirement beyond the identity test.

• Raw whole or cut herb: Macroscopic and microscopic identity, HPTLC fingerprint, foreign matter, heavy metals, pesticide screen, microbial limits, mycotoxins when source or storage indicates risk.
• Powdered herb or blend: Microscopy where possible, HPTLC and HPLC fingerprint, heavy metals, pesticide screen, microbial limits, mycotoxin screen, PAH or pyrrolizidine alkaloid screen where drying method or harvest practice raises risk.
• Standardized extract: HPTLC identity, quantitative HPLC or LC-MS/MS assay, residual solvents, heavy metals, pesticide screen, microbial limits, mycotoxins, targeted adulterant screen for known risks such as marker spiking or synthetic dyes.
• Finished product: Identity of each botanical or its marker per specification, label claim verification, and a contaminant panel scaled to the daily dose, including heavy metals, microbial limits, and pesticides for botanical products.

Reviewing a Certificate of Analysis

A certificate of analysis (CoA) is only as useful as the detail it contains. 21 CFR 111.75(a)(2) permits reliance on a supplier CoA for non-identity specifications only under defined conditions. The CoA must describe the test methods, the limits, and the actual results.

The brand must first qualify the supplier by confirming the supplier's results, document the qualification, and periodically re-confirm it. Quality control personnel must review and approve that documentation.

A CoA suitable for reliance should include the following elements:

• Lot number that matches the material received, with test and issue dates.
• Named method for each test, with a compendial reference or method version.
• Numeric results with units, not just "pass" or "complies."
• Stated limit of quantitation for each contaminant, so that "not detected" has a defined meaning.
• Specification limit next to each result.
• Laboratory name, accreditation status, and an authorized signature.

Common reasons to reject a CoA include a missing lot number, results reported as "<LOQ" with no LOQ value, no method reference, and an identity result based only on a supplier statement. BSCG's guidance on shilajit testing and compliance shows how these principles apply to a high-variability natural material, where detailed heavy metal results and detection limits on the CoA are essential.

Choosing a Laboratory

Request testing from a laboratory accredited to ISO/IEC 17025 for the specific tests performed, and confirm the scope covers botanical matrices. Accreditation demonstrates validated methods, measurement traceability, and controlled reporting.

Other useful questions cover authenticated botanical reference materials for the species in question and proficiency testing for pesticides and metals. Ask also whether the laboratory can perform an orthogonal confirmation when a result is unexpected. Botanical matrices are complex, and a single suspect result should be confirmed by a second method if possible before a lot is accepted or rejected.

Where Third-Party Certification Fits

Third-party certification programs do not replace lot-by-lot botanical testing under GMP. They add independent verification, typically on a defined schedule, and a review of the brand's own quality system and add a marketing component allowing quality to be showcased publicly. Programs that offer some form of independent supplement quality verification include NSF Certified for Sport, Informed Sport, Informed Choice, and BSCG's Certified Drug Free and Certified Quality programs.

The BSCG Certified Quality program states that it verifies ingredient identity and label claims and screens for heavy metals, pesticides, and microbiological agents, along with more than 450 drugs and banned substances. BSCG reports that testing is conducted in ISO/IEC 17025-accredited laboratories using methods including LC-MS, GC-MS, ICP-MS, HPLC, TLC, DNA, and microscopy. It states that limits follow standards from AHPA, AOAC, NSF/ANSI 173, ICH Q3D, USP, Health Canada, EPA, and FDA. The program also reviews supplier qualification and specifications and requires the manufacturer to hold a GMP audit. For ingredient suppliers, the ingredient certification track reviews specifications and quality controls for identity, purity, strength, composition, and contaminant limits.

Programs differ in test frequency, the number of markers verified, contaminant thresholds applied, and whether banned substance screening is included. Brands comparing programs should review program key elements, which BSCG does here in a review of third party certification programs.

Frequently Asked Questions

Is botanical identity testing required on every lot?

Yes, in the United States. 21 CFR 111.75(a)(1) requires at least one appropriate identity test or examination on every lot of a dietary ingredient before use. A supplier CoA cannot substitute for this test. The only alternative is a petition under 21 CFR 10.30 for reduced testing, supported by scientific data and carried out on terms FDA specifies when it grants the petition.

Can DNA barcoding be used for botanical extracts?

DNA barcoding works best on raw or lightly processed plant material where DNA is intact. Extraction with heat and solvents degrades or removes DNA, so a negative DNA result on an extract does not prove the plant is absent. Compendial reviews recommend chemical fingerprinting methods such as HPTLC and HPLC for extracts, with DNA as a supporting method for raw material only.

Which heavy metals should botanical testing cover?

Arsenic, cadmium, lead, and mercury are the four elements addressed by USP General Chapter <2232> for dietary supplements. Some programs and laboratories add chromium or other elements based on risk. Results should be evaluated against the daily dose of the finished product, since limits are expressed as permitted daily exposure rather than concentration alone.

How do pesticide requirements differ between the United States and the European Union?

The United States relies on EPA tolerances and USP General Chapter <561> limits, which do not cover every pesticide and botanical combination. The European Union applies Regulation (EC) 396/2005, which sets a default maximum residue level of 0.01 mg/kg wherever no specific level exists. Brands selling in both markets generally specify the stricter limit and confirm the laboratory method can quantify at that level.

Do standard contaminant panels detect adulteration?

No. Heavy metal, pesticide, and microbial panels do not detect substitution with another plant, spiking with a purified marker compound, or undeclared synthetic drugs. Detecting adulteration requires identity methods that distinguish the genuine material from known adulterants, plus targeted LC-MS/MS screens for drugs or stimulants where the ingredient category carries that risk.

What should a botanical CoA include for a brand to rely on it?

Under 21 CFR 111.75(a)(2), the CoA must describe the method used, the limits applied, and the actual results. In practice this means a matching lot number, numeric results with units, a stated limit of quantitation, a compendial or validated method reference, and the laboratory's identity and accreditation. The brand must also have qualified the supplier by confirming its results and must periodically re-confirm. A CoA that reports "<LOQ" without stating the LOQ value, or that omits the lot number, does not meet this standard.

Summary

Botanical testing for supplement ingredients rests on four pillars. These are orthogonal identity confirmation on every lot, a validated strength assay for any numeric claim, a contaminant panel scaled to the material and daily dose, and targeted adulterant screens where risk is documented. Identity testing is the only element fixed by regulation for every lot. The remaining tests depend on specifications the brand sets using compendial chapters, association guidance, retailer requirements, and destination-market law.

Two distinctions matter most. Contaminant testing and adulterant testing answer different questions and require different methods. A supplier CoA supports, but does not replace, the brand's own verification. Numeric limits cited here reflect specific standards and regulations at the time of writing; the source documents hold the current values.