SkinSort vs INKEEDecoder: the same ingredient numbers, two different answers
In our 20-product study, SkinSort and INKEEDecoder gave the same irritancy and pore-clogging rating to an ingredient 101 times out of 101. They still disagreed on whether four products were fragrance-free or alcohol-free. Here is why, and when to use each.
In our 21 October 2026 study of 20 products, INKEEDecoder (formerly INCIDecoder) and SkinSort gave identical irritancy and pore-clogging ratings in 101 of 101 shared ingredient pairs, yet disagreed on a fragrance-free or alcohol-free tag for four products, through a different rule or a different formula on file.
What is the difference between SkinSort and INKEEDecoder?
They share the numbers and differ in what they do with them:
Same ingredient ratings: 101 of 101 irritancy and 101 of 101 pore-clogging ratings matched where both rate the same ingredient.
INKEEDecoder explains, it does not score: no product score, an ID-Rating per ingredient (superstar, goodie, icky) and a few hashtag highlights.
SkinSort sorts: about a dozen yes or no tags per product and, after sign-up, a personal Match Score out of 100.
They part on four of 20 products, because of a fragrance rule or because each holds a different version of the formula.
Why do two sites with the same numbers give different verdicts?
Because a verdict is a rule applied to a list, and the rule or the list can differ while every number matches. Drunk Elephant Protini is the clean example of a rule: both sites hold the same 58-ingredient list, and INKEEDecoder even labels linolenic acid and phenylpropanol "perfuming" in its function column, yet it tags the cream "#fragrance & essentialoil-free" while SkinSort marks it not fragrance-free. On 21 October 2026 SkinSort counted both ingredients as fragrance; on 23 October 2026 it counts phenylpropanol alone. A third database, Skincarisma, sides with INKEEDecoder: "Fragrance-free" at a glance, on a 58-ingredient list that includes phenylpropanol. Neutrogena Hydro Boost and Laneige Lip Sleeping Mask are examples of the list: the two sites hold different formulas, so one finds an alcohol the other does not.
Same inputs, different outputsthree products, re-read 23 October 2026
"Shared ratings" counts ingredients both sites rate for irritancy and pore-clogging in that product (study, 21 October 2026). SkinSort's Hydro Boost page opens on its UK listing, which carries 38 ingredients including t-butyl alcohol; INKEEDecoder's list has 20. INKEEDecoder's Laneige list includes "Alcohol", which SkinSort's 34-ingredient list does not.
How do SkinSort and INKEEDecoder compare feature by feature?
INKEEDecoder is a reference; SkinSort is a shopping tool. INKEEDecoder, which renamed itself after the Personal Care Products Council, which runs the INCI naming system, filed a trademark lawsuit over "INCI" (the dispute "has since been resolved"), keeps "the same website, the same ingredient database". It gives every ingredient a plain-English "what-it-does" line and refuses a product score. SkinSort adds what a shopper wants: tags you can click to see the triggering ingredient, a Match Score from Poor (0 to 59) to Perfect Match (95 to 100), an iOS app with seven free scans, and a Premium tier with unlimited scans and ad-free browsing.
Feature matrixfrom each site's own pages, read 23 October 2026
INKEEDecoder's founder discloses on its about page that she co-owns the Hungarian webshop Skinsmart.hu and co-founded the brand Geek & Gorgeous. SkinSort's about page says it "has never taken outside funding", and its disclosure says links may be affiliate links. In-app prices are from the US App Store.
Where do they disagree across the 20 study products?
On tags for four products, and on the ingredient list itself for 10. The fragrance-free tag splits once (Drunk Elephant Protini); the alcohol-free tag splits three times (LRP Toleriane Double Repair, Neutrogena Hydro Boost, Laneige Lip Sleeping Mask). The ingredient count differs for 10 of 20, usually by one or two names, but by 18 for Hydro Boost and 12 for Laneige. The first column is the reason to trust either site's per-ingredient numbers: every pair matched.
INKEEDecoder against SkinSort20 products from the 21 October 2026 study
the two sites agree
they differ
no rated ingredient in common
Fragrance-free: INKEEDecoder's "#fragrance & essentialoil-free" highlight against SkinSort's fragrance-free attribute. Alcohol-free: INKEEDecoder's "#alcohol-free" highlight against SkinSort's alcohol-free attribute. Ingredient count: INKEEDecoder vs SkinSort. Vaseline is a single ingredient that neither rates.
Which should you use, SkinSort or INKEEDecoder?
Both, for different jobs, and neither as a safety verdict. Use INKEEDecoder to understand a formula: what each ingredient is for, and where its numbers come from (its comedogenicity explainer warns the rabbit-ear data are mostly undiluted and "The concentration matters a lot"). Use SkinSort to shop: filter by tags, compare dupes and, if you sign up, sort by a Match Score that SkinSort itself cautions has limits: "there's only so much that we can infer from ingredient lists" (SkinSort support). Before you trust either tag, match its ingredient list to your box. For how these two sit among the other checkers, see the best ingredient checker app and who pays the ingredient apps.
Using both sites on one product
1. Open both pages and compare the ingredient count
with your box first: a mismatch means an old or
regional formula, not a disagreement.
2. Trust the per-ingredient numbers on either site:
they matched in 101 of 101 pairs we checked.
3. On SkinSort, click a "Not ..." tag to see which
ingredient triggered it.
4. On INKEEDecoder, read the icky ingredients and the
"what-it-does" column, not just the hashtags.
5. Both rate ingredients, not doses; neither is a
medical opinion.
Who should pick which?
If you want to understand an ingredient listINKEEDecoderPlain-English functions per ingredient, and no product score to argue with.
If you shop with filtersSkinSortAbout a dozen yes or no tags, each showing the ingredient behind it.
If you avoid fragrance strictlySkinSort's tag, then checkIt flagged Protini where INKEEDecoder did not; read which ingredient it counts.
If you scan on your phoneSkinSortINKEEDecoder has no app in the US App Store; SkinSort gives seven free scans.
If the two disagreeCount the ingredients10 of 20 products had different lists; a different formula explains most splits.
How we compared them
Method
The 20 products, chosen before any site was read, are those of our 21 October 2026 ingredient-app study. For each product we matched ingredients both sites rate and compared their irritancy and comedogenicity numbers (101 pairs), then compared INKEEDecoder's highlights with SkinSort's attributes and each site's ingredient count. Protini, Hydro Boost and Laneige were re-read on 23 October 2026; SkinSort's Protini fragrance count had changed from two ingredients to one, and its tag had not. Features come from each site's own pages and the US App Store, read 23 October 2026. This compares what the sites publish; it is not a test of any product.
As an Amazon Associate I earn from qualifying purchases.
FAQ
Is INCIDecoder the same as INKEEDecoder?
Yes. INCIDecoder renamed itself INKEEDecoder after the Personal Care Products Council, which runs the INCI naming system, filed a trademark lawsuit in the US; the site says the dispute has been resolved and that the website and ingredient database are the same. The incidecoder.com address still works.
Do SkinSort and INCIDecoder use the same ingredient ratings?
In our 21 October 2026 study they matched every time: 101 of 101 irritancy ratings and 101 of 101 comedogenicity ratings were identical where both sites rated the same ingredient in the same product.
Why does SkinSort say a product is not fragrance-free when INKEEDecoder says it is?
Usually a rule. For Drunk Elephant Protini, SkinSort counts phenylpropanol as a fragrance ingredient (read 23 October 2026) while INKEEDecoder tags the cream fragrance and essential-oil free, though it lists phenylpropanol and linolenic acid as "perfuming".
Does INKEEDecoder have an app?
Not in the US App Store: a search for INKEEDecoder or INCIDecoder on 23 October 2026 returned no app from it. An app called InciDecode is from an unrelated developer. SkinSort has an iOS app with seven free scans.
Is SkinSort free?
Partly. A free account gives seven product scans in the app; Premium adds unlimited scans, a Match Score breakdown, sorting by match and ad-free browsing. US App Store in-app prices run from $7.99 to $59.99.
Which is more accurate, SkinSort or INKEEDecoder?
On per-ingredient ratings they agree exactly. Where their tags differ, the cause in our study was a different fragrance rule or a different formula on file, so check each site's ingredient list against your product before trusting either tag.
Bottom line
SkinSort and INKEEDecoder are built on the same ingredient numbers and differ in what they build on top: INKEEDecoder explains, SkinSort sorts. When they disagree, it is a rule or a formula version, not the chemistry, so read the list against your box and use whichever answers your question.