Data Sources
Not every fragrance's data comes from the same place. AuraNote is upfront about which tier applies to any given fragrance — the same labels shown in the app appear here.
1Where the catalogue comes from
Almost none of the catalogue is written by AuraNote. Of the 47,515 fragrance pages published today, 45,331 of them (95%) are built from an imported Fragrantica-derived corpus: the brand, name, release year, gender, accords, note pyramid, rating and vote count all originate there. A smaller, older set of imports comes from a Parfumo-derived CSV. What AuraNote computes on top of that — note prominence, families, scent twins, dupes — is described in Methodology, and it is derived from those imported fields rather than measured independently.
This is worth stating plainly because the rest of this page describes how a fragrance's performance figures are sourced, and it would otherwise read as though everything else were AuraNote's own.
2Curated fragrances
About 150 fragrances are hand-reviewed: their note pyramids, performance figures, and descriptions are authored directly rather than computed. These don't carry a data-source label in the app — their values are treated as authoritative.
3Community data
"Exist" is a specific test rather than a judgement call: the fragrance needs at least 20 raters, and both a longevity average and a sillage average must be present. Anything short of that — five real raters, or raters who scored one dimension but not the other — is labeled "Estimated from notes" below, not "Community data".
4Estimated from notes
5How to check for yourself
Open any fragrance's page and look at its performance card — the label next to Longevity and Sillage tells you exactly which of these three tiers applies to that specific fragrance.