Methodology
How the data on this site is sourced, checked, and kept honest.
1. Facts, not borrowed prose
A fragrance's note pyramid, release year, perfumer and concentration are facts. They are not copyrightable and they are published here as structured data. The descriptive text on each entry is composed by Perfume Codex from that structured data — we do not copy descriptive writing from other reference sites.
2. Photographs
Every image on a fragrance entry is an original render produced by Perfume Codex in a single consistent style: pure white background, consistent framing and our own watermark. We do not hotlink or republish manufacturer or third-party photography. Member-contributed photographs are accepted only under a licence that permits us to display them.
3. Batch codes
Batch code rules are published with a confidence level and a source link. A batch code tells you approximately when a bottle was filled. It is not a proof of authenticity — counterfeiters print codes too. We state this on every batch code page.
4. What we publish, and what we hold back
An entry is published only when we hold a minimum factual dataset for it: a complete three-layer note pyramid, at least three recorded accords, and a release year. Entries that do not meet that threshold are held back rather than padded out. This is deliberate: a reference library is measured by the reliability of what it contains, not by the number of pages it serves.
5. Corrections
Every entry carries a revision history. If you find an error, tell us and we will correct it and record the change. Corrections are not a sign of weakness in a reference work; silent inaccuracy is.