The Recommended For You(RFY) widget delivers personalized product suggestions by learning from each shopper’s unique browsing and purchase history. As a dynamic “shopping assistant,” RFY adapts in real time—surfacing relevant items on the homepage, search‐no‐results pages, product detail pages, cart pages, and beyond.
Recommended Placement

1. Place on the homepage to greet returning shoppers with tailored picks.

2. Add to PDPs or zero-results pages to reengage users when no organic results match.

​
How It Works

The following logical order would be used by the algorithm to generate RFY recommendations:
  1. Collaborative Filtering
    Leverages patterns across your user base—“shoppers who viewed X also viewed Y.”
  2. Content Signals
    Analyzes product attributes (categories, tags, descriptions) to surface items with similar properties.
  3. Session-Based Signals
    Prioritizes products the user has interacted with during their current session.
  4. Similar Items Fallback
    If session data is sparse, defaults to algorithmically detected similar products.

​
Supported Rule Types

You can scope RFY behavior at multiple levels, each of which can override the global settings:
  • Global (default for all placements)
  • Home Page
  • Product Page
  • Category Page
  • Cart Page
  • Search Page

​
Behavior for Not Logged-In Users

  • With Session Activity
    RFY tracks clicks/views in the current session and populates suggestions accordingly.
  • New Visitors
    • Fallback Disabled: Widget remains hidden.
    • Fallback Enabled: Displays products from your chosen fallback algorithm.

​
Behavior for Logged-In Users

  • Returning Customers
    RFY sequentially applies collaborative, content, then session logic to craft highly relevant suggestions.
  • Freshly Registered (No History)
    • Fallback Disabled: Widget remains hidden until they interact.
    • Fallback Enabled: Shows fallback recommendations to kickstart engagement.
Fallback settings are a UI-level feature. The underlying AI/ML API returns an empty result if no recommendations can be generated.

​
Example Scenario

  1. Context: A shopper lands on a designer-furniture PDP for a mid-century sofa.
  2. Action: RFY analyzes that user’s past views/purchases of coffee tables and accent chairs.
  3. Outcome: A “Recommended For You” carousel appears beneath the product details—showing complementary pieces (side tables, cushions) that match their style.