The Similar Products algorithm identifies items that share key characteristics (category, brand, price range, attributes) or user-behavior patterns with the currently viewed product. By surfacing visually or semantically related alternatives, it enriches the shopping experience, helps shoppers discover new options, and drives cross-sell opportunities.
Recommended Placements

1. Product Page : Display “Similar Products” alongside the main product details to offer alternatives or complementary items.

2. Cart Page : Recommend products similar to those already in the shopper’s cart.

3. Checkout Page : Suggest related add-ons or accessories before final purchase.

​
How It Works

  1. Feature Extraction
    Products are represented by a combination of metadata (category, brand, price, attributes) and latent factors learned from user interactions (views, purchases).
  2. Similarity Scoring
    A similarity score is computed between the seed product and every other item in the catalog using a hybrid of metadata matching and collaborative signals.

​
Supported Rule Types

  • Global
  • Home Page
  • Product
  • Cart

​
When to Use

  • Alternative Discovery
    Help customers find comparable or upgraded versions of the item they are considering (e.g., a different color, style, or price tier).
  • Complementary Upsell
    Surface accessories or companion products that share attributes with the main product.
  • Inventory Balancing
    Promote in-stock items similar to out-of-stock products to reduce bounce rates.

​
Example

A shopper lands on a product page for a “Mid-Century Modern Leather Sofa.”
  1. Experro’s Similar Products algorithm computes similarity across sofas in the same category, brand, and price range.
  2. It ranks alternatives like “Vintage-inspired Leather Loveseat” and “Rustic Leather Sectional” based on combined metadata and past shopper behavior.
  3. The page displays these options under “You May Also Like”—encouraging the shopper to explore complementary or upgraded styles.