Ingredient Scanner
Get the iOS app

Methodology

How the information here is made.

Where the data comes from, how a name on a label is matched, how articles are written and checked, and what none of it can tell you.

The data

The app and this site share one ingredient dataset, version 2026.10.10.2. It holds 34,785 ingredient names from CosIng, the European Commission’s database of cosmetic ingredients, together with the names labels commonly use for them, each ingredient’s functions, and the EU rules that apply to it. The data is selected, reorganised and normalised, never rewritten: a function or a restriction on this site is the one the source records.

CosIng is an information database with no legal value, and a name appearing in it does not mean the ingredient is approved for use in cosmetics. Nothing in the dataset is a safety assessment.

How a name on a label is matched

The app, and the ingredient checker on this site, sort every name they read into one of these:

  • Recognised: the exact ingredient name, or the name printed with a note beside it, such as Tocopherol (Vitamin E).
  • Known alias: another name the source records for the same ingredient, such as Glycerine for Glycerin.
  • Corrected: a label quirk or a misreading with one possible answer, such as two names printed without a comma, a plant part left out, different spacing, or a character a camera mistakes for one that looks like it, such as a zero for the letter O. A spelling is corrected only when every difference is such a look-alike, exactly one ingredient fits and no number changes. Shown with the text as printed, and can be undone.
  • Suggested match: anything less certain, such as a missing letter, a different number, or a name with several meanings like Vitamin E. Up to three readings are offered, the likeliest first, and none is applied until you choose.
  • Unknown: not in the dataset. It stays on the list as unknown and is never treated as harmless.

Roles and groups

An ingredient’s roles are its official EU functions, with a plain-language phrase for each. Where it has several, the most telling comes first: a preservative is named as one before its other jobs. A substance from a fragrance group leads with its scent, unless it is also a preservative. On the haircare and body care pages the role that kind of product uses it for leads, as in the app: Behentrimonium Chloride conditions hair, Zinc Ricinoleate reduces odor.

A group, such as Fragrance or Sulfates, is a fact about an ingredient: it belongs to a chemical family or to a list in EU law. People use groups to find what they prefer to avoid. Being in a group is not a judgement of the ingredient.

What we never do

  • Give a product a safety score or a grade. An ingredient list has no amounts, and the formula changes how each ingredient behaves.
  • Call an ingredient toxic, clean, safe or unsafe on its own.
  • Tell anyone a product suits their allergy, condition or pregnancy.
  • Rate or rank brands and products, or accept payment for placement.

How articles and guides are written

There are two kinds of ingredient page. 500 dictionary entries are built from the dataset, with a short plain-language summary; they are useful when you arrive from a label, and they are kept out of search results until they are rewritten as full articles. 18 in-depth articles and 12 guides are written by hand, and only those are offered to search engines.

  • Sources, in order of weight: EU law and the opinions of the EU’s Scientific Committee on Consumer Safety; US FDA guidance; published safety assessments such as the Cosmetic Ingredient Review; then peer-reviewed research. Other ingredient websites are never used as sources.
  • Every claim is cited. Anything that is not common knowledge carries a numbered link to its source, and every source was opened and checked against the sentence citing it on the date shown.
  • Who writes them. Articles are drafted with the help of AI writing tools, then checked against their sources before they are published. They have not been reviewed by a dermatologist or a toxicologist; if that changes, the reviewer will be named on the page.
  • Corrections. If a page is wrong or out of date, tell us with the page and, if you can, a source. Each page shows when it was last checked.

What none of it can tell you

A label lists ingredients, not amounts; the order is only roughly by amount, and below 1% not at all. Parfum can stand for many unnamed scent ingredients. Text recognition can misread a label. Nothing on this site or in the app is a safety, allergy or medical assessment. If you react to a product, speak to a doctor or pharmacist.

Sources and licences

Ingredient names, functions and descriptions: CosIng, the European Commission database of cosmetic ingredients (https://ec.europa.eu/growth/tools-databases/cosing/), retrieved 2026-10-09. Common ingredient names: Commission Implementing Decision (EU) 2025/1175. Regulatory notes: Annexes II to VI to Regulation (EC) No 1223/2009 as exported by CosIng (last updated 06/10/2026). Historical names: CosIng inventory of 15/02/2016. Harmonised classifications: Regulation (EC) No 1272/2008, Annex VI (consolidated 20260701). E-numbers: Regulation (EC) No 1333/2008, Annex II (consolidated 20260818). Household ingredients: US EPA Safer Chemical Ingredients List. Compound identifiers: PubChem, NCBI. © European Union, 1995-2026, reused under Commission Decision 2011/833/EU. This app selected, reorganised and normalised this data; CosIng is an information database without legal value, and nothing here is a safety assessment.