Adobe Commerce only. Learn more.

Create New Recommendation

When you create a recommendation, you are creating what is called a recommendation unit:

Recommendation unit

The products inside the recommendation unit are called recommended items.

After you create and activate the recommendation unit, Adobe Commerce will begin collecting metrics that measure data such as impressions, views, clicks, and so on. The Product Recommendations dashboard displays these metrics for each recommendation unit to help you make informed business decisions.

  1. On the Admin sidebar, go to Marketing > Promotions > Product Recommendations to display the Product Recommendations dashboard.

  2. Specify the Store View where you want the recommendations to display.

    You can create recommendation units for the Page Builder page type only in the default store view.

  3. Click the Create Recommendation button.

  4. In the Name your Recommendation section, enter an internally-facing name that uniquely identifies the recommendation on your dashboard, such as Home page most popular.

  5. In the Select page type section, select the page on which you want the recommendation to appear.

    You can create up to five active recommendation units for each page type. If a page type option is grayed-out in the Admin UI, the limit has been reached and an explanation is shown. However, the Page Builder page type allows you to create up to 25 recommendation units because you can use Page Builder to edit arbitrary content types.

    Options are:

    • Home Page
    • Category
    • Product Detail
    • Cart
    • Confirmation
    • Page Builder
  6. In the Select Recommendation type section, specify the type of recommendation you want to appear on the selected page type. Note that certain page types restrict what type of recommendation you can select.

  7. In the Storefront display label section, enter an externally-facing name your shoppers will see, such as “Top sellers”.

    Adobe Commerce provides the following guidance for label names.

    If you selected Page Builder as your page type, you will specify the Text or Heading element as your storefront display label in the Page Builder editor.

  8. In the Choose number of products section, use the slider to specify how many products you want to appear in this recommendation unit.

    The default is 5. Maximum is 20.

  9. In the Select placement section, specify the container where you want the recommendation unit to appear on the page.

    Options are:

    • At the bottom of main content
    • At the top of main content
  10. (Optional) Select and move the rows in the Choose position table to order the recommendations within the container.

    The Choose position section displays all recommendations (if any) created for the page type you selected.

  11. (Optional) In the Filters section, you can apply filters that control which products can be displayed in recommendations.

  12. When you finish configuring your recommendation you can click:

    • Save as draft to edit the recommendation unit at a later date. Note that you cannot modify the page type or recommendation type for a recommendation unit in a draft state.

    • Activate to enable the recommendation unit on your storefront.

Preview Recommendations

Use the Recommended products preview panel to preview what products will be displayed in a recommendation unit when deployed to your storefront.

Recommendation preview

The preview panel supports fetching recommendations from a different environment. This allows merchants who are testing in a non-production environment to experiment with rules and preview the recommendations before deploying to production.

Field Description
Name Name of the product
SKU Stock Keeping Unit of the product
Price Price of the product
Result Type Primary indicates that there is enough training data collected to display a recommendation. Backup indicates there is not enough training data collected so a backup recommendation is used to fill the slot. See Training time for machine learning models to learn more about machine learning models and backup recommendations

As you create your recommendation unit, experiment with the page type, recommendation type, and filters to get immediate real-time feedback about the products that will be included. As you begin to understand what products will be displayed, you can configure the recommendation unit to best meet your business needs.

What appears in the preview panel and on the storefront can occasionally vary. This is because Adobe Commerce filters recommendations to avoid displaying duplicate products when multiple recommendation units are deployed on a single page.