Methodology
How we review the research
Every number in a Stack Report traces back to a primary source that a credentialed human reviewed. Here is exactly how evidence gets in, how it is graded, and why the score cannot be bought.
Ziad Meras, M.Sc. Organic Chemistry
Founder & reviewer · Last reviewed 2026-07-20
The review pipeline
- 1Papers are entered into a source registry — one record per study, before anything is extracted.
- 2AI extracts dose ranges, bioavailability, and interaction data from the source text. At this stage it is never used in scoring.
- 3A credentialed reviewer checks each extracted record against the primary source, corrects errors, and marks it human-reviewed. Only reviewed records feed the score.
- 4When enough reviewed records exist for a compound, a scoring parameter (range + evidence tier) is generated and confirmed by the reviewer.
Your report is calculated by deterministic scoring software against our reviewed research database — it is not written by an AI language model. That database is built by extracting study data from primary sources with AI, then verifying it against those sources by a credentialed reviewer; your free-text entry is matched to compounds by deterministic text-matching, and you confirm those matches before anything is scored. This is not a substitute for professional medical advice.
Evidence Tiers (A / B / C / D)
Every compound-level finding shows its Evidence Tier. The tier is derived mechanically from the underlying studies, so it is auditable rather than a judgment call.
- Evidence Tier A, StrongStrong
≥1 meta-analysis or ≥2 independent human RCTs, adequate combined sample, relevant population.
- Evidence Tier B, ModerateModerate
A single human RCT, or multiple consistent cohort studies.
- Evidence Tier C, LimitedLimited
Observational/cohort only, or animal studies with mechanistic plausibility.
- Evidence Tier D, PreliminaryPreliminary
In-vitro or animal-only evidence, or a single small human study.
How evidence bounds the score
The Spend Efficiency Index is capped by evidence tier: strong dosing on a weakly-evidenced compound cannot produce a high score. The ceilings below are fixed and applied to every scored compound.
Why the score is independent
Your Spend Efficiency Index is calculated independently of any affiliate relationship. Products we link to may earn a commission — this never affects your score or evidence ratings.
The scoring engine and the product-recommendation engine are separate systems with no shared ranking logic; commission data is not an input the score can see.