Glossary of scoring terms.

Overall score

A single 0-10 number combining six weighted pillars.

The composite is a weighted sum of six sub-scores: Dose vs. studies 50%, Transparency 20%, Value 10%, Formulation 10%, Safety 5%, and Community 5%. Every input is public. No brand can pay to change any input. Read the full formula on the methodology page.

Evidence grade (A-D)

How strong the human trial evidence is for an ingredient.

A = multiple large RCTs (N>500) or meta-analyses confirm efficacy. B = human RCTs with biomarker outcomes. C = early human pilots or a single RCT. D = preclinical or mechanism only. Grades apply to ingredients, not products, and only upgrade when new human evidence is published. Downgrades require editorial review.

Evidence tier (T1-T5)

How direct the evidence is for the exact product or protocol.

T1 = RCT tested the exact product SKU. T2 = RCT tested the same ingredient at the same dose. T3 = RCT tested the ingredient at a different dose. T4 = human observational or n=1 self-experiment with instrumented tracking. T5 = mechanistic hypothesis, animal or in vitro only. Blueprint self-reports sit at T4.

Trial dose

The daily dose used in the human trial we cite.

For each ingredient we pin one reference dose from the most relevant published human RCT (or, when several exist, the median dose across trials). Products are scored on how close their labelled daily dose sits to this reference. Trial doses are shown on every ingredient page next to their PubMed link. Not to be confused with the brand's recommended serving.

Dose vs trial dose (%)

Labelled daily dose as a percentage of the trial dose.

100% means the product delivers exactly the dose the human trial used. Above 100% is over-dosing (only useful if the ingredient has a wide safety window); below 100% is under-dosing (fine for some ingredients, harmful to the score for most). Blends are capped at 45% per ingredient so a single trace ingredient cannot carry the whole product.

RxD-$ (per trial dose day)

Cost per day at the actual trial dose, not the label serving.

If a product is labelled 500 mg but the trial used 1000 mg, RxD-$ is calculated at two capsules per day. This lets you compare a small bottle at a small price against a large bottle at a large price on the same axis: what you would actually pay to reproduce the trial. Shown in USD.

PRIMARY

The main active ingredient the product is scored on.

Every product has one PRIMARY ingredient plus zero or more secondary ingredients. The Dose vs studies score is based mainly on the PRIMARY (weighted at up to 100% for single-ingredient products, capped at 45% for blends). Shown as a green tag next to the ingredient name in every ingredient list.

CoA (Certificate of Analysis)

A lab report showing what a specific batch actually contains.

A Certificate of Analysis is a third-party or in-house lab test verifying the amount of active ingredient in a specific production batch, plus heavy metals, microbial contamination, and residual solvents. A brand publishing recent, batch-specific CoAs is one of the strongest signals of transparency. We check whether the CoA is public and current per product.

Tested by independent lab

Verified by an independent lab, not the manufacturer.

Third-party testing means samples are sent to an independent lab (NSF, USP, Informed Choice, Eurofins, Alkemist) which has no financial tie to the brand. In-house testing does not count. Tested by independent lab products get a bonus in the Transparency and Safety scores. Certification programs like NSF Certified, USP Verified and Informed Choice add another bonus on top.

NDI (New Dietary Ingredient)

A US FDA notification required for ingredients not sold before Oct 15, 1994.

Under DSHEA, any dietary ingredient not marketed in the United States before Oct 15, 1994 requires a New Dietary Ingredient notification to the FDA at least 75 days before commercial launch. FDA can object; if it does, the ingredient cannot be sold as a supplement. NMN is currently in an unresolved NDI status in the US (July 2026).

EU Novel Food

An EU authorisation required for foods not consumed to a significant degree before May 15, 1997.

Regulation (EU) 2015/2283 requires that any food or food ingredient not consumed to a significant degree within the EU before May 15, 1997 must be pre-authorised by the European Commission before it can be sold. Applications are assessed by EFSA. NMN received an EFSA positive opinion in May 2026 but has not yet been listed by the Commission, so it remains unauthorised in the EU.

GRAS (Generally Recognized As Safe)

A US FDA status for food ingredients with adequate safety history.

GRAS is an FDA determination that a substance has been sufficiently proven safe under the conditions of its intended use. It can be self-affirmed by the manufacturer (with an independent expert panel) or notified to the FDA with no objection response. Nicotinamide Riboside received GRAS status in 2017. GRAS applies to food use, not to any specific health claim.

Bayesian shrinkage

A statistical correction that pulls small-sample ratings toward the average.

A product with 5 reviews at 4.9 stars is less trustworthy than one with 5000 reviews at 4.6 stars. Bayesian shrinkage adjusts each rating toward a prior (we use 3.5 stars with 20 pseudo-reviews) so a small number of very high ratings is not treated as strong evidence. The result: a highly-rated product with lots of reviews scores better than a highly-rated product with few reviews.