Scentum

Interpretive perfume discovery for a generation that doesn't trust marketing copy.

5D Vector

A taste engine disguised as a shop

Scentum is not e-commerce. It's a translation layer between structured fragrance data and human perception — turning notes, accords, and ratings into a navigable aesthetic space. Every perfume lives as a 5-dimensional vector, not a single 1-to-100 score.

  • Authenticity — natural vs synthetic feel
  • Projection — intimate vs loud
  • Longevity — lasting power on skin
  • Complexity — diversity of notes & evolution
  • Versatility — how widely it wears
The ScoreVector — five dimensions: Authenticity, Projection, Longevity, Complexity, Versatility AUTHENTICITY PROJECTION LONGEVITY COMPLEXITY VERSATILITY

Why a vector, not a score

A single 1-to-100 rating collapses fragrance into a hierarchy. But "best" depends on whether you want to be noticed, to be remembered, or to disappear into your own coat. Scentum maps every perfume into five orthogonal dimensions — so the question stops being "what's good?" and becomes "what's you?"

  • Authenticity ratio of natural to synthetic feel
  • Projection from skin-scent intimate to room-filling loud
  • Longevity hours on skin before it's gone
  • Complexity how much it evolves between top and base
  • Versatility how many moods and seasons it survives

Three pillars

The interpretive layer is built from three loosely-coupled systems: a vector model, a vibe taxonomy, and an AI discovery flow that grows on top of both.

Pillar I
ScoreVector

  • Five orthogonal dimensions per perfume
  • Derived heuristically from notes & ratings
  • Refined offline by Claude Haiku enrichment
  • Drives filters, similarity, and recommendations
  • Visualised as radar, bars, or interpretive prose

The substrate everything else stands on. Five clean dimensions that voice and conversation can modify fluidly.

Pillar III
AI Discovery

  • Faceted filters as the base layer
  • Natural-language search via Claude
  • Voice input for full-sentence intent
  • Conversational refinement loop
  • Personalised "Scent Story" for each match

The AI is a co-pilot that progressively narrows possibility space through dialogue — search becoming conversation becoming guidance.

Vibe Tags

Concrete examples of the cultural vocabulary the catalog speaks. Each tag is a cluster of perfumes that share a feeling, not a formula.

quiet-luxury romantic archive-core clean experimental late-french-theory linen-on-skin burnt-sugar-saturday rainy-bookshop first-frost salt-and-cedar midnight-amber

Under the hood

A small, dependable pipeline. Public web data, an offline enrichment step, and a frontend that treats the database as a vocabulary, not a stockroom.

Scentum data pipeline: Scrapy → Zyte → Claude → MongoDB → React/TS

Scrapy + Zyte

Two-phase spider on Zyte Cloud crawls multiple fragrance sites, deduplicating across runs to surface only newly published listings.

Claude Enrichment

Haiku derives the missing fields — authenticity, complexity, versatility, vibe tags — from notes, accords, and reviewer language.

MongoDB Catalog

Document store holding each perfume with its full ScoreVector, ready for faceted queries and similarity ranking.

React / TypeScript

Mobile-first interface with semantic sliders, vibe chips, AI-augmented search, and an interpretive product page.

From search to conversation to guidance

The four-step shift from user-driven search to AI-as-co-pilot. Each step is independently shippable; together they compose a fragrance consultant, not a search box.

1

Voice as Input

Speech-to-text feeds raw utterances. Claude maps them to the ScoreVector and explains the mapping — interactive, not opaque.

2

Fragrance Oracle

An optional 3-to-5 question guided flow. Each answer animates a live ScoreVector radar — results feel earned, not browsed.

3

Refinement Loop

"Too sweet", "warmer", "less woody" — quick reactions adjust weights live. Preference learning inside a single session.

4

Scent Story

The product page becomes a mirror — fragrance described back through the user's own language and revealed memories.

The interpretive layer is live

Scentum is in active development at scentum.space. The catalog grows every scrape; the AI discovery layer ships in increments. Try it, break it, and tell me what feels off.