Once you have defined your Recommender types, you can use the Auto recommender dashboard to create and manage the actual recommendation rules. This feature allows you to dynamically suggest specific “Target” products based on the “Source” products a customer is currently viewing or interacting with.

Auto Recommender Dashboard

The main grid provides a clear overview of all your active and inactive recommendation rules. From here, you can:

• Create an Auto recommender: Use the Create Auto recommender button at the top right of the screen to build a new rule.
• Filter & Search: Use the Filter menu to easily find specific rules, and reset the view using the Clear all button.
• Manage rules: The grid lists key information for each rule, including its Name, Priority, Enable status, the Localized catalog(s) it applies to, and the specific Recommender Type it is Applied to. You can click the Edit link in the Actions column to modify an existing rule.

Creating and Updating an Auto Recommender

When you create or edit an auto recommender, you will configure its behavior through a detailed form separated into general settings and rule engines.

General Settings

• Enable: A toggle switch to quickly activate or deactivate the rule on your storefront.
• Name: A mandatory, human-readable label for your rule (for example, “Tops and Bottoms”).
• Localized catalog(s): A multi-select field allowing you to define exactly which catalogs and store views this recommendation applies to.
• Applied to: This mandatory dropdown links your rule to a structural Recommender type (such as the “Default”, “Up-sell”, or “Cross-sell” types you created previously).
• Active (From / To): Date pickers allowing you to schedule time-bound recommendation campaigns.
• Priority: A mandatory numeric value. If multiple auto recommenders match a given product context, the system will use this priority score to determine which rule takes precedence.

Rule Engines

The core logic of your recommendations is built using two distinct visual condition builders:

• Source rule engine: This defines the triggering context. You build conditions here to identify the product the user is currently looking at.
  • Example: Set a condition where Category is equal to Bottoms. The rule will only trigger when a user views pants, shorts, etc.
• Target rule engine: This defines the recommended products. You build conditions here to determine which items will be pulled from your catalog and displayed to the user.
  • Example: Set a condition where Category is equal to Tops. The system will fetch shirts, jackets, etc., to recommend.

Both the Source and Target engines allow you to combine complex criteria, giving you the flexibility to require that all conditions are true, or that any conditions are true, ensuring highly relevant cross-merchandising.