Decades of reproducible clinical evidence for real changes to skin physiology. The ingredients dermatology reaches for first.
Retinoids (retinol, retinal, tretinoin), L-ascorbic acid, proven UV filters, well-studied AHAs/BHAs.The Skincare Ingredient Analyzer That Scores Results. We Don't Just Read Labels — We Audit Them.
Most ingredient checkers stop at safety. DemythSkin asks the harder question first: does this actually work? The engine reads the full INCI list and grades the formula against clinical evidence on two separate axes — so you can see past the marketing on the front of the bottle to the chemistry inside it.
How The Analysis Works
Retrieve the real formula
Scan the barcode or type the name. The analyzer identifies the product and retrieves its exact INCI list — the full, legally ordered ingredient lineup. Incomplete lists are rejected rather than scored, and formulas that look outdated get flagged.
Judge the chemistry, not the brand
The engine ignores the brand name, the price, and the packaging claims, and evaluates the formula the way a skeptical cosmetic chemist would: what's in here, how much of it, and what does the published evidence say it does?
Run the fixed rubric
Every ingredient is mapped against a database of 3,000+ compounds. A deterministic rubric — not an AI's opinion — computes the efficacy and safety scores, plus a better-value alternative where the product falls short.
How The Analyzer Scores Products
Does it actually work?
- Performance over marketing: An ingredient earns its score from clinical evidence of what it does on skin — never from the claim on the box.
- Concentration matters: A proven active does nothing at a token amount. Position in the INCI list works as a concentration proxy, so we penalize "fairy dusting" — a trace of a hero ingredient added purely to name-drop it on the label.
- Chemistry & stability: Good ingredients still fail in a bad formula. We check what decides whether actives actually function: is the pH right for exfoliating acids? Does the packaging shield unstable antioxidants like vitamin C from light and air? Is it a studied active, or a cheaper derivative standing in for one?
- Judged against its own job: A cleanser is scored on cleansing without stripping — not on actives that rinse off in thirty seconds. A sunscreen is scored on filter coverage and photostability. Each product type gets the rubric that fits it.
Will it irritate?
- Severity, not headcount: Deductions scale with the evidence. Added fragrance high in a formula, drying alcohol as a main solvent, or a harsh sulfate as the primary surfactant count heavily. A declared trace allergen counts a little. Five trace allergens are not treated like one stripping face wash.
- Evidence-gated flags: An ingredient only reaches the irritant tier if it's documented in patch-test literature or carries a regulatory restriction — and occurs at meaningful frequency. Rare sensitizers get a quiet note, not a warning banner.
- No preservative panic: Standard preservatives, emulsifiers, and thickeners are shown as formulation base and scored as neither good nor bad. They keep your product stable and uncontaminated; treating them as hazards is fear, not science.
- Pore-clogging visibility: Known comedogenic ingredients — heavy waxes and pore-clogging esters — are always surfaced in the breakdown, so acne-prone users can see them rather than hunting the label.
EFFICACY GOLD STANDARD 4.5+ · EFFECTIVE 3.9–4.4 · MODEST 3.5–3.8 · MID 3.0–3.4 · UNDERPOWERED 2.5–2.9 · WEAK 2.0–2.4 · INEFFECTIVE below 2.0
SAFETY VERY GENTLE 4.5+ · GENTLE 3.9–4.4 · MOSTLY GENTLE 3.5–3.8 · SOME RISK 3.0–3.4 · IRRITATING 2.5–2.9 · HARSH 2.0–2.4 · AVOID below 2.0
The Ingredient Tiers
Solid clinical backing for barrier repair, hydration, and skin health. The dependable workhorses of good formulation.
Ceramides, niacinamide, hyaluronic acid, glycerin, panthenol, squalane, peptides, tocopherol.Real but developing evidence — genuinely useful, not score-carrying on their own.
Centella compounds, green tea, allantoin, botanical antioxidants, emollient plant lipids.Heavily promoted, thin clinical support for the headline claim. Not harmful — just not what's driving results, so never credited as if it were.
Topical collagen, gold, caviar, diamond powder, most "stem cell" extracts.Established sensitizers and irritants — admitted only with patch-test literature or a regulatory restriction behind them, at meaningful frequency.
Fragrance/parfum, declared EU allergens, essential oils, denatured alcohol, formaldehyde releasers, harsh sulfates.Solvents, preservatives, emulsifiers, thickeners, and pH adjusters. These make a product stable and usable — they aren't what makes it work or not work, so they add nothing to efficacy and count against safety only in the rare, documented cases.
Water, phenoxyethanol, carbomer, xanthan gum, disodium EDTA, cetearyl alcohol, dimethicone.Every product in both is scored by the same rubric above — nothing sponsored, nothing paid for placement. Support the project →