How we evaluate fragrances

ScentCriticis synthesis criticism: scores do not come from one person's nose but from crossing thousands of public wear reports, composition data and our own database of fragrances compared against each other. Every score comes out of that research, and this page explains exactly how, so you can decide how much to trust it.

How we evaluate

For each perfume we aggregate the sources that actually contain experience: community vote distributions (longevity, sillage, season), buyer reviews at retailers, documented note pyramids and accords, published perfumer interviews, and the reformulation and batch history. On top of that we run our own model: every fragrance is compared against the rest of our database (similarity percentages, performance gaps, value by price tier) and the result becomes scores plus an editorial verdict.

When we truly test, and how we mark it

When we wear a fragrance on skin (usually via decant), the entry carries the "Hands-on tested" badge, which states how many days and when; only then will you read first-person experience, and we always distinguish which observation comes from our own wear and which from the aggregated consensus. Our editorial system physically blocks any wear claim about a fragrance without that record: the badge is not decoration, it is the technical precondition for writing "we wore it", so if you do not see it, everything you read comes from the measured consensus, not our wrist.

What we do NOT do

We never fake experience: if a fragrance does not carry the tested badge, everything you read about its performance comes from the measured consensus of the people who wear it, not our wrist. We also do not take bottles from brands in exchange for coverage, and no agreement with any store touches a score. When the community splits on a scent, we report the split and explain it instead of hiding the fight behind an average.

Our scoring

The overall score runs from 0 to 10 and is an editorial judgement informed by the data, not an automatic vote average. Alongside it we publish longevity, sillage, projection, versatility and value on a 1 to 5 scale, derived from the aggregated consensus. Skin chemistry varies, and we say so when it matters.

Corrections

When a figure changes or we get something wrong, the affected article is corrected and says so: its update note records what changed and when, and the "catalog data verified" date on every piece shows the last automated check against our database. If you spot an error, write to us through the contact page and we will review it.

Who signs

Every article is signed: Álvaro Cendra does the analysis and Noelia Ibarra reviews it before publication: their check verifies that every figure resolves to the catalog and that no wear claim exists without a testing record. Their author pages list everything they have signed.

How we're funded

We never show a fixed price, because it may vary: the actual amount is confirmed at the store at the moment of purchase.