How we review products
Show the math. Explain the evidence. Name the reviewer.
Our proposed methodology combines evidence, safety, value, and individual fit. It is designed to make buy, buy-if, and skip verdicts equally useful.
Version 1.0 · September 2026. This is our intended operating standard. Team assignments, credential checks, and medical reviews are still pending.
What enters the catalog
A product must disclose its full ingredient list, avoid demonstrably false claims, and pass an initial safety screening. Listing is not endorsement. Brands cannot pay for inclusion, exclusion, or removal.
A transparent score out of 100
The proposed overall score is 0.35 × evidence and formulation + 0.25 × safety and tolerability + 0.20 × value + 0.20 × fit. Each subscore is measured on a 0–100 scale.
- Evidence and formulation · 35%. Human evidence for the claimed concern, effective concentrations, and a plausible delivery system.
- Safety and tolerability · 25%. Irritation and sensitization risks within category norms, with relevant safety and pregnancy flags.
- Value · 20%. Cost per effective dose against comparable formulas.
- Fit · 20%. Suitability for intended skin types and concerns. Targeted products are assessed against their intended audience.
Evidence by concern
The planned A, B, and C evidence tiers apply to each concern, not to an entire product. Citations must be checked against the journal record. Missing evidence is labeled “early evidence.” Detailed tier thresholds and the evidence table must be finalized before this scoring system is used.
Real-world outcomes
Verified-purchaser reports from independent retail sources will be considered alongside research. Brand-only reviews indicate low-confidence data. Minimum sample thresholds and the criteria for an “outcome-backed” label must be published before that label is used. Disagreement between trials and buyer reports should be explained, not hidden in an average.
The signed verdict
Every completed product review is intended to carry a two-to-four-sentence buy, buy-if, or skip verdict written or countersigned by a named, verified dermatologist. It should include skin-type and condition fit, routine placement, potential conflicts with other products, and cheaper equivalents when available.
What readers can inspect
The overall score, four subscores, evidence tiers and sources, data confidence, full ingredient list, reviewer identity, and all flags. Methodology changes require a public changelog. Version 1.0, September 2026: foundation proposed; operational scoring criteria pending.
Methodology foundation
Our foundation credits DermApproved as an influence on transparent scoring, concern-specific evidence tiers, and verified citations. Our proposed approach adds named reviewer accountability and individual fit. This does not imply an affiliation or endorsement.
Historical ratings are separate
The cleanser advertorial reproduces the original guide’s scores out of 10 and letter grades. They are not our proposed scores out of 100, evidence tiers, independent testing, or a completed medical review. No original product scoring has been completed in this preview.