How to read a Fragrantica entry without falling into traps

A vast collective memory

Fragrantica is a perfume encyclopedia, a magazine and a community, launched in 2007 by Elena and Zoran Knezevic. It is an independent site, not affiliated with any fragrance house, and it now holds well over a hundred thousand perfume entries, thousands of listed perfumers and millions of user reviews. At that scale, no other public resource offers a comparable collective memory of perfumery.

That breadth is its first strength. To find the stated launch year of a release, the note list a brand has communicated, or the verdict of hundreds of wearers on longevity, Fragrantica has no real rival. The site also runs an editorial side, distinct from the community database, with interviews, trade-fair reporting and bylined columns.

An entry, though, blends three kinds of information that do not carry the same weight: what the brand declared, what the community perceived and voted, and what stays uncertain for lack of a source. Telling them apart is what separates a useful reading from a copied error. This piece is not a complaint about Fragrantica. It is a way to use it with discernment.

The pyramid, what the brand declares and what the community votes

The fragrance pyramid, top, heart and base notes, is the most consulted part of an entry. On Fragrantica it usually mirrors the list communicated by the brand or its press office. It is declarative data, not an independent analysis of the juice in the bottle.

Next to the pyramid, the site shows main accords as bars, fed by community votes. That layer is valuable for grasping a perfume's general impression, but it is prone to a well-known bias, the power of suggestion. A wearer who reads the word vanilla before smelling tends to vote vanilla. The most voted accords are therefore not an objective measurement but an average of impressions nudged by what is on screen.

A simple habit helps. Highly specific notes, a Florentine iris pallida or an incense of a stated origin, almost always come from a press kit and deserve trust. Generic notes, amber, woods, white musk, may reflect marketing copy as much as collective perception. The more specific and checkable a note is, the more seriously it can be taken.

The practical takeaway is clear. The pyramid tells you what the brand claims. The voted accords tell you how the public reads it. Neither tells you what the perfumer actually built. For that you need a source closer to the creation: a press release, a brand dossier or an interview with the nose.

Perfumer credits, what is documented and what is missing

Fragrantica credits perfumers when the information is available, usually supplied by the brand or the trade press. A perfumer's page lists the compositions attributed to them. The database is useful, but it has two limits worth knowing.

First, attribution depends on what a brand is willing to disclose. Many houses keep their perfumers anonymous, and the entry then reads as unknown author or stays silent. Second, some attributions circulate without a primary source and end up copied elsewhere. An attribution read on an entry is only a certainty once a brand or a documented interview confirms it.

The rule we apply is firm: a perfumer credit enters an Osmetheca entry only if it is confirmed beyond the community database alone. The same principle protects any reader from a misattribution.

Release years, reliable at launch and trickier afterward

The year shown on an entry is, in principle, the launch year. For the major landmarks it is generally reliable. The gaps appear elsewhere, and they are common.

Three cases call for caution. Flankers, whether intense, extrait or limited editions, sometimes carry a different year from the original, and the confusion is easy. Reformulations do not change the displayed date even though the juice has changed. And for old or confidential releases, the year sometimes rests on a community estimate rather than a brand source.

Here too, the entry is a good entry point, not proof. A date that matters, in an article or a reference page, deserves to be cross-checked against the house's official site, a press release or the specialist press.

How we cross-check a Fragrantica entry

The Osmetheca method comes down to one rule: no fact, date, attribution or figure is published without being triangulated across at least three converging editorial sources. Within that method Fragrantica has a precise, useful place, never the place of a single source.

In practice, a Fragrantica entry serves to spot a lead, compare a note list or gauge a collective perception. It is then set against sources closer to the creation: the house's official site first, press kits, institutions such as the Societe Francaise des Parfumeurs or the Osmotheque de Versailles, and then independent criticism. When those sources converge, the information is solid. When they diverge, we flag it rather than rule arbitrarily.

A concrete example makes it tangible. To date a perfume, you note the year shown on the entry, then open the house's official site looking for a launch mention or a press release. If the two agree, the date is settled. If they diverge, you add a third source, specialist press or an institution, and keep the majority version while flagging the doubt. The routine takes a few minutes and prevents most of the errors that travel from one site to the next.

When to rely on it, when to complete it

When can you lean on Fragrantica without reservation? To explore a catalogue, discover neighboring perfumes, read hundreds of opinions on longevity or sillage, and follow the editorial coverage of a fair or a house, the site is excellent and hard to replace.

When should you complete it? As soon as a fact has to be asserted: the exact launch year, a perfumer's name, the precise composition of a reformulated juice or an ingredient's lineage. On those points the entry is a start, and confirmation comes from elsewhere.

Read well, a Fragrantica entry does real work: it opens leads and takes the pulse of a community. Read badly, it spreads approximations that harden as they are copied. The difference is not in the site. It is in how you use it.

See also