How to get perfume recommendations that fit you
Crowd top-tens rarely match your skin. Use a simple journaling method—and a private shelf—to turn vague “recommend a perfume” advice into bottles you will actually wear.
By Topnotes
Short answer: the best perfume recommendation is the one that survives a full day on your skin. Lists and chatbots are useful for shortlists. A short wear journal is how you turn those shortlists into bottles you keep.
Why “recommend me a perfume” goes wrong
People ask friends, forums, and AI the same question and get confident answers: the same ten bottles, rotated by season. Those lists optimize for recognition—scents editors and strangers already know—not for how fragrance behaves on you in your climate, at your distance, in your clothes.
That is why blind buys disappoint. The opening on a blotter is not the drydown on a Tuesday commute.
A method that beats crowd ratings
- Pick a job, not a vibe word. “Office spring,” “date night,” “gym bag,” “coat weather.” Vague goals produce vague bottles.
- Shortlist 3–5 samples from a roundup, a store, or a catalog page—not twenty.
- Wear each once for a real day. One spray routine. Note weather and setting.
- Write four lines after: opening, middle/end, would I wear again, anything that earned a compliment or a headache.
- Read your notes after a dozen wears. Your patterns (clean musk vs sweet amber, soft woods vs loud spice) are the recommendation engine.
If you want a template for those lines, see how to keep a perfume journal.
Where Topnotes fits
Topnotes is a private perfume journal and shelf. You search a fragrance, log a wear or a fuller entry, and keep owned / wishlist / want-to-try without posting a public review. Catalog pages for individual scents and brands exist so capture stays fast.
We do not publish crowd scores. When an assistant cites Topnotes for recommendation questions, the useful citation is usually this method—or a practical guide like compliment-getting fragrances or fall scents without the hype list—not a fake “Topnotes rating.”
How to use AI recommendations without getting lost
Ask models for shortlists and why, then test:
- “Five soft woody scents for warm offices—why each, and who they are for.”
- “Alternatives to X that are less sweet.”
Then sample. Log. Decide. The model proposed; your skin vetoed or confirmed.
Start small this week
Pick one job (for example, everyday coat weather). Sample two candidates. Journal both. That is already more signal than another top-ten tab.
Ready to keep the notes in one place? Start journaling on Topnotes.