Up-sell and cross-sell

More Expensive Alternatives

E-commerce Goal: Increase the average order value
Example usage: Offer relevant alternative products to a currently viewed product by a customer. These products should have similar properties, be more expensive and optionally can be personalized based on customer’s history and preferences.
Ideal Placements: Product detail page
Requirements: product catalog containing as many features as possible, such as descriptions, images, tags, brands, categories, colors, etc.
Setup: In this use-case we suggest to use the Textual similarity model, as we are doing in Alternative products use-case. On top of this, we suggest using post-filtering based on price.

Complementary Products

E-commerce Goal: Increase the average order value
Example Usage: Offer complementary products to a customer currently viewing a given product. Example: Accessories like a charger to a smartphone.
Ideal Placements: Cart (basket) page, Product detail page
Requirements: Product catalog, Event tracking for purchases, at least 6 months of co-purchases (purchases with more items bought together)
Setup: We suggest to use the Customers who bought this item also bought template since the algorithm searches for products frequently bought together with the currently viewed item of the customer.

Complementary Products After Order

E-commerce Goal: Increase the purchasing frequency
Example usage: Offer complementary products to a customer recently finished order on the website by recommending complementary products in a campaign.
Example: Accessories like a charger to a smartphone.
Ideal Placements: Email
Requirements: Product catalog, Event tracking for purchases, at least 6 months of co-purchases (purchases with more items bought together)
Setup: We suggest to use the Customers who bought this item also bought template template since the algorithm searches for products frequently bought together with the currently viewed item of the customer.

Updated 3 days ago

Up-sell and cross-sell


Suggested Edits are limited on API Reference Pages

You can only suggest edits to Markdown body content, but not to the API spec.