Skip to main content

Scent Intelligence for your CRM

Traditional CRMs react to what people buy. WikiParfum helps you understand why they buy — turning olfactive preference data into hyper-personalized communications, deeper emotional connections, and better customer retention. WikiParfum links all preferences and interactions to a stable user identifier. This means the olfactive intelligence built during a website session doesn’t end when the user leaves — it can be reused across channels, sessions, and over time.

The business impact

  • Higher open rates and click-through — messages that resonate with personal scent preferences outperform generic promotions
  • Better retention — identify patterns to reconnect with customers before they churn
  • Smarter upsell and cross-sell — suggest complementary products based on each customer’s olfactive profile
  • Stronger brand perception — customers feel understood and valued

What you can build

  • CRM enrichment — append olfactive preferences (favorite families, ingredient affinities) to user profiles
  • Personalized email campaigns — recommend fragrances in newsletters based on each user’s taste profile
  • Lifecycle journeys — trigger re-engagement with new launches that match a user’s olfactive preferences
  • Taste-based segmentation — group users by olfactive profile for targeted marketing
  • Cross-channel consistency — same recommendations on web, mobile, in-store, and email

How it works

The foundation is the session identity established via initSession (see Sessions & User Identity).
1. User visits your site → initSession with userId
2. User explores, views perfumes, interacts with ingredients
3. WikiParfum captures these as olfactive preference signals
4. On next visit (or in email/CRM) → same userId retrieves evolved preferences
Because preferences are linked to a stable identifier, taste-based personalization remains relevant even as product assortments evolve.

CRM integration

WikiParfum enhances your existing CRM — no replacement, no heavy setup, no data migration. The connection is anonymous: WikiParfum never sees or stores your customer data. All personal information stays within your system. Compatible with: Salesforce, Adobe Campaign, Bloomreach, Emarsys, Insider, Klaviyo, HubSpot, and more.

User-scoped recommendations

Pass userId to any recommendation query to get results influenced by the user’s accumulated interaction history.
query PersonalizedRecommendations {
  recommendedPerfumes(
    userId: "4ad05833-01e9-412f-b6fd-a02c2ce1f7cd"
    search: { lang: "EN", perfumes: "424" }
  ) {
    id
    name
    brand {
      name
    }
    family {
      name
    }
  }
}
Over time, recommendations become increasingly aligned with each user’s evolving taste profile — without requiring explicit questionnaires or profile setup.

CRM integration patterns

Newsletter personalization

Use recommendation data to populate email templates with fragrances tailored to each recipient:
  1. Retrieve the user’s userId from your CRM
  2. Call recommendedPerfumes with recent interactions or catalog additions
  3. Include the recommended perfumes (with images and family data) in the email template

New launch targeting

When a new fragrance launches, identify which users are most likely to enjoy it:
  1. Get the new perfume’s olfactive profile (family, ingredients)
  2. Match against users whose preference history aligns with that profile
  3. Target those users with launch communications

Segmentation by olfactive profile

Group users by their dominant family affinities or ingredient preferences for targeted campaigns — for example, sending woody fragrance promotions to users who consistently engage with woody perfumes.

Cross-session persistence

The same userId works across:
  • Multiple website visits — preferences accumulate over time
  • Different devices — as long as the user is identified
  • Different touchpoints — web, mobile app, in-store kiosk, email
This enables a consistent understanding of taste that improves with every interaction, regardless of where it happens.