ScentyAISign in
Stop blind-buying fragrance.
Tell us what you love and what you cannot stand. We will tell you which five fragrances to sample next, why each fits you — and, just as plainly, what might make you dislike it.
Takes about two minutes. Nothing is saved until you choose to create an account.
Every pick shows both sides
Most recommenders only sell you the upside. The reason you waste money is the thing nobody told you. So a ScentyAI recommendation is never allowed to show a fit note without its friction note, in the same size and the same weight.
- Fit note
- Warm tonka and cinnamon over a dry woody base — very close to the Khamrah you rated highly, but drier and far less sweet.
- Friction note
- Built on clean white musk. Around one in thirteen people of European ancestry carry a receptor variant that makes this musk class smell like almost nothing.
Grounded in smell research
Two people differ functionally at more than 30% of their olfactory receptor genes. That is the real reason reviews disagree — and why we tell you when a fragrance leans on a note that a large share of people genuinely cannot smell.
Similarity measured the way the field measures it
Whole-profile angle distance, not counting shared notes — the approach that actually predicts perceived similarity in the literature.
Confidence we can defend
Predicting one person’s taste is genuinely hard, so we show a band and say what the uncertainty comes from rather than inventing a percentage.
122 fragrances, designer to niche to Middle Eastern
Chosen for accuracy over volume, each with a confidence rating you can see.