Where Quality Programs Break Down

Most supplement and cosmetic manufacturers have some form of incoming raw material testing. The problem is not usually the absence of testing — it is the absence of structure. Testing happens, but specifications are vague, sampling plans are undocumented, and the connection between test results and lot release decisions is unclear.

Under 21 CFR Part 111 (dietary supplements) and ISO 22716 (cosmetics GMP), incoming material testing is a regulatory requirement, not a best practice. But the regulations give manufacturers significant latitude in how they design their programs. That latitude is only useful if you fill it with a documented, defensible structure.

This post walks through the key components of an incoming raw material testing program: specifications, sampling plans, test protocols, and the decision framework for lot acceptance and rejection.

Component 1: Raw Material Specifications

A raw material specification is the document that defines what an acceptable lot looks like. It is the standard against which all test results are evaluated. A specification that is too vague cannot be used to make defensible lot release decisions; a specification that is too narrow will generate unnecessary failures.

What a specification must include:

1. Identification information

Material name (common name, INCI name, CAS number, or botanical nomenclature as applicable)
Supplier name(s) — if you accept the same material from multiple suppliers, consider whether the specification should be supplier-specific
Grade or quality level (food grade, pharmaceutical grade, USP grade, etc.)

2. Physical and chemical parameters Each parameter should include:

The test method (by name and number — e.g., “Moisture: USP <921> Loss on Drying”)
The acceptance criterion (e.g., “NMT 5.0%”)
The unit of measurement

Typical parameters for botanical raw materials include: appearance, odor, particle size, moisture/loss on drying, ash content, heavy metals, pesticide residues, and marker compound assay. For mineral ingredients: appearance, assay, heavy metals, and dissolution rate. For lipid-based ingredients: appearance, assay, peroxide value, acid value, and moisture.

3. Identity test Every specification must include at least one identity test. The identity test should be capable of distinguishing the material from likely substitutes or adulterants. For botanicals, HPTLC with a reference standard is the minimum; DNA barcoding should be added for high-risk species.

4. Microbial limits Microbial acceptance criteria should be consistent with USP <2023> for the material’s intended use (oral supplement, topical cosmetic, etc.). For raw materials that will be used in products without a terminal sterilization step, microbial limits are particularly important.

5. Labeling and documentation requirements The specification should list the documents required with each shipment: CoA, SDS, allergen declaration, GMO declaration, country of origin certificate, and any other material-specific documentation.

Component 2: Sampling Plans

A sampling plan defines how many containers to sample, how much material to take from each container, and how to composite or retain samples. Without a documented sampling plan, your test results cannot be interpreted as representative of the lot.

Statistical basis for sampling

The most commonly used sampling plan reference for pharmaceutical and supplement applications is the ANSI/ASQ Z1.4 standard (attribute sampling) or the approach described in USP <1> (Injections) and related chapters. For raw material incoming testing, a simplified approach based on the square root of N plus one (√N + 1) is widely used:

Count the total number of containers in the lot (N)
Sample √N + 1 containers, rounded up to the nearest whole number
For very large lots (>100 containers), a fixed sampling frequency may be appropriate

This approach is a practical starting point. For high-risk materials or materials with known homogeneity issues, a more rigorous sampling plan may be warranted. Consult your quality team on the appropriate plan for each material category.

Sample size

The sample size per container should be sufficient to perform all required tests plus retain a reference sample. For most solid raw materials, 100–200 grams per container sampled is adequate. For liquids, 100–200 mL. Retain at least one sample from each lot for the duration of the product’s shelf life plus one year.

Sampling procedure

Document the sampling procedure in a standard operating procedure (SOP). The SOP should specify:

Sampling equipment (stainless steel thief, ladle, pipette — appropriate to the material)
Decontamination procedure for sampling equipment
Container opening and resealing procedure
Sample labeling requirements
Chain of custody from sampling to laboratory

Component 3: Test Protocols

A test protocol is the document that specifies which tests to perform on a given material, in what order, and what to do with the results. It translates the specification into an actionable testing workflow.

Tiered testing approach

Not all tests need to be performed on every lot. A tiered approach allocates testing resources based on risk:

Tier 1 (every lot):

Identity (HPTLC or equivalent)
Appearance/organoleptic evaluation
Moisture/loss on drying
Microbial limits (TAMC, TYMC, specified organisms per USP <61>/<62>)

Tier 2 (every lot from new suppliers; every 3rd–5th lot from qualified suppliers):

Assay/potency (HPLC or equivalent)
Heavy metals (ICP-MS per USP <232>/<233>)
Pesticide residues (for botanical ingredients)

Tier 3 (annually or upon supplier re-qualification):

Full specification testing including all Tier 1 and Tier 2 parameters
Additional parameters specific to the material (e.g., solvent residues, mycotoxins)

The tier assignment for each material should be documented in the specification and reviewed when supplier qualification status changes.

Method references

Every test in the protocol must reference a specific, validated method. “Heavy metals by ICP” is not a method reference. “Heavy metals: ICP-MS per USP <232>/<233>, Method QC-HM-001, version 2.1” is a method reference.

For tests performed by a contract laboratory, the protocol should reference the laboratory’s method identifier and confirm that the method is validated for your specific matrix.

Component 4: Lot Acceptance and Rejection

The lot acceptance decision is the output of the incoming testing program. It should be documented, traceable, and made by a qualified person.

Acceptance: All required tests pass against specification. Lot is released to quarantine storage pending manufacturing use. Release documented in the batch record with test results, CoA reference, and authorized signature.

Conditional acceptance: One or more parameters are outside specification but within an established deviation range. Requires documented investigation, risk assessment, and quality management approval before use. Not appropriate for identity failures.

Rejection: Any identity failure, or any result outside specification that cannot be resolved through investigation. Lot is segregated, labeled “REJECTED,” and returned to supplier or destroyed per your disposition procedure. Rejection documented with root cause investigation.

Inconclusive results: When a test result is inconclusive (e.g., HPTLC shows an unusual pattern that does not clearly pass or fail), the lot should be held pending additional testing. Do not release a lot with unresolved inconclusive results.

Building an Incoming Raw Material Testing Program — Specifications and Protocols