Glo’s scoring engine is built on established cosmetic chemistry, regulatory science, and peer-reviewed dermatology. No proprietary black boxes. Here’s what we draw from.
We don’t maintain a single proprietary database. We aggregate from multiple established sources and cross-reference them against each other so no single dataset’s blind spots become ours.
Peer-reviewed clinical studies on ingredient safety, efficacy, and dermatological outcomes. The primary evidence base for active ingredient claims.
The Scientific Committee on Consumer Safety publishes detailed opinions on cosmetic ingredient safety, concentration limits, and restrictions.
Concentration limits, banned substance lists, and sunscreen monograph data from the FDA and international cosmetic regulators.
The International Nomenclature of Cosmetic Ingredients. We parse every label using INCI conventions, then translate to common names in 11 languages.
An open-source, community-maintained database of cosmetic product ingredient lists. Used for barcode scan lookups, supplemented by our own OCR pipeline.
Published research from the Journal of the American Academy of Dermatology, British Journal of Dermatology, and Contact Dermatitis, among others.
Every scan follows the same five-step pipeline. The entire process takes under two seconds.
You point your camera at a product label or barcode. For images, we extract the full INCI list using vision AI. For barcodes, we look up the product in Open Beauty Facts.
Every INCI name is parsed, separated, and translated to its common name. “Butyrospermum Parkii Butter” becomes “Shea Butter.” Original order preserved — position matters for concentration.
Each ingredient is classified (good, okay, warning, flagged) based on its safety profile, comedogenicity, regulatory status, and relevance to your skin.
Up to ten goal-specific sub-scores are calculated based on your profile. Sensitivity and skin type carry the highest weight. The Skin Match Score is a weighted average of all applicable goals.
Both scores are delivered alongside plain-English explanations: what’s good, what’s flagged, and why. Every ingredient gets a status. No vague labels.
The scoring engine evaluates each ingredient against known research. Here are the key categories we track.
Retinoids, niacinamide, vitamin C (ascorbic acid, MAP, SAP), AHAs, BHAs, PHAs, peptides, ceramides, hyaluronic acid, azelaic acid, arbutin, kojic acid, tranexamic acid, and bakuchiol.
Fragrance/parfum, essential oils, SLS/SLES, high-concentration acids, menthol, camphor, witch hazel, alcohol denat., and formaldehyde releasers.
12 known comedogens scored on a 0–5 scale: coconut oil, cocoa butter, isopropyl myristate, wheat germ oil, lanolin, and others. Weighted by INCI position.
Ceramides, cholesterol, fatty acids, phytosphingosine, niacinamide, panthenol, and centella/madecassoside. We track multi-layer moisture architecture.
Retinoids, high-dose salicylic acid (>2% BHA), hydroquinone, certain chemical sunscreens, formaldehyde, and phthalates. Flagged when pregnancy is in your profile.
Ingredients banned or restricted by EU or FDA regulations, concentrations exceeding established safety limits, and substances under active regulatory review.
We take transparency seriously. That includes being honest about our limits.
Glo is a research tool, not a medical device. Our scores don’t replace professional dermatological advice. If you have a skin condition, see a doctor.
Our barcode database covers approximately 92% of products in major markets. For the other 8%, we rely on OCR from the label photo.
Brands rarely disclose exact concentrations. We infer relative amounts from INCI position (listed in descending order) and known typical ranges.