Skip to main content

​
What It Is

The Enriched Product Recommendation Use Case delivers individually-selected product suggestions within email communications. Unlike static batch recommendations, these are computed at the moment of open using the freshest available data — last browsing session, recent purchases, current stock levels, and individual preferences. The “enriched” aspect means going beyond collaborative filtering: recommendations incorporate multiple data dimensions (affinity, seasonality, stock, offers, lifecycle stage) to select the most relevant products for each individual at that exact moment.

​
Key Challenges Addressed

This Use Case addresses the broadest range of challenges — 21 Key Challenges across 7 categories:
CategoryChallenges addressed
Customer Lifecycle & ValueFirst purchase, one-timer conversion, frequency, basket increase, category expansion, churn reduction, winback, low-spender, VIP conversion, reduce promo dependence
Loyalty ProgrammeProgramme recruitment, status progression, points utilisation
New Data UsageProprietary scoring, intent qualification, preference collection
Language ManagementLanguage adaptation, real-time personalisation
Marketing PressureAutomatic rotation after non-conversion
App DownloadApp-exclusive pricing display
OmnichannelCoherent product suggestion across channels

​
Required Datasources

DatasourceFields usedRole
CRM 360Customer ID, purchase history, preferences, segmentIndividual targeting
Product CatalogueSKU, name, price, image, category, attributesProduct selection pool
NavigationViewed products, categories, time spentReal-time intent signals
Purchase HistoryTransaction date, products, amountsAffinity and exclusion logic
Stock & AvailabilityStock level, availability by locationNever recommend out-of-stock
OffersCurrent promotions, eligibilityPrice display with offers applied

​
Required Capabilities

CapabilityRole in this Use Case
Recommendation EngineSelects products using collaborative filtering + individual signals
Lifecycle Offer EngineApplies the right offer to the recommended product based on lifecycle stage
Category AffinityIdentifies preferred categories for cross-sell / category expansion
Language ScalingRenders product details in the individual’s locale
Currency AdaptationShows correct pricing for the individual’s market

​
Example Implementation

​
Workflow Logic

Data Node: CRM 360 → retrieve customer segment, preferences, locale
Data Node: Product Catalogue → retrieve eligible products
Data Node: Navigation → recent browsing (last 7 days)
Data Node: Stock → filter in-stock only

Logic Node (Branch):
  IF customer.segment = "one-timer"
    → Output Node: "Complete your look" — recommendations from same category as first purchase
  ELSE IF customer.last_purchase_days > 60
    → Output Node: "We miss you" — top-affinity products with exclusive offer
  ELSE IF customer.segment = "VIP"
    → Output Node: "Selected for you" — new arrivals in preferred categories
  ELSE
    → Output Node: "You might like" — general recommendations weighted by affinity

​
What the Customer Sees

A personalised product block within the email showing 3–6 products selected specifically for them, with:
  • Product image, name, and price (in their currency and language)
  • Any applicable offer (loyalty points, promotion) shown
  • Products they already purchased excluded
  • Products currently in stock at their preferred store/region

​
Measuring Incremental Value

MetricHow to measure
Click-through rateExposed vs Non-Exposed on recommendation block
Conversion ratePurchases from recommended products within 7 days
Average basketCompare basket size for buyers who clicked recommendations
Revenue per emailTotal revenue attributed to recommendation clicks
See Key Challenges — Frequency and AOV for valorisation formulas.