How to Read an NMN Certificate of Analysis

The direct answer

Start by matching the document to the product. A certificate for bulk NMN powder can report the purity of that raw material, but it cannot show how much NMN is in a finished capsule. A finished-product report can answer that question only when the product name, variant, serving basis, and lot match the bottle being evaluated.

Then read the result as a complete line: analyte, method, specification or acceptance limit, actual numeric result, and unit. A large purity percentage, a pass mark, or an HPLC label means little when the sample and measurement basis are unclear.

Raw material and finished product answer different questions

Raw-material identity: Is the incoming ingredient the substance the supplier says it is?
Raw-material purity: What share of the tested ingredient sample was the target compound?
Finished-product potency: How much NMN was measured in the tested capsule, serving, or unit?
Finished-product contaminants: Did the tested lot meet the listed limits for the selected contaminant panel?

A 99.9% raw-material purity result is not the same as 500 mg in a finished serving. Purity is a proportion of the tested ingredient sample. Potency or strength is the measured amount in the finished unit. Encapsulation, blending, serving size, and other ingredients sit between those two stages.

A facility registration, GMP certificate, laboratory accreditation, or supplier qualification record is also not a product result. Those documents may describe a system or organization, but they do not provide the measured NMN amount for a particular finished lot.

Check these seven fields in order

Sample: exact ingredient, product, variant, and dosage form tested.
Lot or batch: the number that connects the report to a production run.
Laboratory and date: who performed the test and when the sample was analyzed or reported.
Analyte: NMN, beta-NMN, a contaminant, or another compound actually measured.
Method: the analytical method or method description used for that measurement.
Specification: the target, range, maximum, minimum, or other acceptance limit.
Actual result and unit: the measured number and whether it is mg, mg per capsule, percent, ppm, CFU, or another unit.

US dietary-supplement manufacturing rules provide a useful benchmark for document content. Under 21 CFR 111.75, a manufacturer relying on a supplier's certificate of analysis must first qualify that supplier. The certificate must include the test or examination methods, test limits, and actual results. The rule also requires appropriate, scientifically valid methods when determining whether specifications are met.

That regulation does not say every consumer-facing product page must publish every COA. Public disclosure is an editorial quality signal because it lets a reader inspect the product connection; it should not be mislabeled as a universal publication requirement.

Match the lot before reading the number

A result is most useful when the lot on the report matches the lot printed on the bottle or the current product page clearly identifies the tested lot. If no lot appears, the report may still describe a product or ingredient, but it cannot establish which production run the number belongs to.

Dates need context too. The manufacture date, sampling date, analysis date, report date, and expiration or retest date are different fields. A recent webpage linked to an old certificate does not make the test recent. Record the date shown on the report itself.

Product names must match beyond the brand. A certificate for an NMN 1200 formula cannot support a different NMN 300 product simply because both come from the same company. Different serving sizes, capsule counts, added ingredients, and delivery formats can represent different finished products.

Read method, limit, result, and unit together

HPLC is a broad analytical technique, not a guarantee that every report is equivalent. The method must be suitable for the analyte and sample matrix, and the report should make clear what was measured. A result such as 513.21 mg per capsule answers a different question from 99.91% purity.

Pass or conform can be useful when the acceptance limit is visible, but a numeric result provides more information. For example, “lead: pass” cannot be interpreted without the method, reporting unit, and maximum limit. “Not detected” also needs a detection or reporting limit before it can be compared across reports.

Units are part of the result. Milligrams per capsule, milligrams per serving, percent by weight, parts per million, and micrograms per day cannot be treated as interchangeable. Compare like with like and convert only when the serving basis and units are explicit.

Contaminant panels have a defined scope

Heavy metals, microorganisms, residual solvents, and other contaminants require separate tests and limits. A report that lists arsenic, cadmium, lead, and mercury does not automatically cover pesticides, microbes, or residual solvents. Record which panels were actually performed.

21 CFR 111.70 requires finished-batch specifications for identity, purity, strength, composition, and limits on contamination that may adulterate the supplement. Under 21 CFR 111.75, manufacturers may verify a statistically selected subset of finished batches, or every finished batch, as part of the production and process control system. This is why one public report should be described as evidence for its tested sample and lot, not every bottle ever made.

What a certificate of analysis cannot prove

It does not prove that NMN produces a clinical benefit.
It does not show that a higher labeled dose is appropriate for a person.
It does not make a raw-material result a finished-product result.
It does not turn laboratory accreditation into product approval.
It does not cover contaminants that were not included in the test panel.
It does not support a different variant, serving size, formula, or lot.

After reading the document, use our NMN comparison method to normalize dose and price. The NMN human trials guide covers clinical evidence, while our five-product review shows how raw-material and finished-product reports changed the rankings.