Install and Configure Recommendations

Deploying Product Recommendations to your storefront requires that you install, configure, then create the recommendations in the Admin UI.

Install Product Recommendations

Because the Product Recommendations module is a stand-alone metapackage, updates are released more frequently than Magento Commerce. To make sure you are up-to-date with the latest bug fixes and features, refer to the release notes.

Install the magento/product-recommendations module with Composer:

1
   composer require magento/product-recommendations

The magento/product-recommendations module requires the following dependencies:

• data-services — This module enables behavioral data collection by tracking user events on the page. This type of data is required by Adobe Sensei to compute product affinities based on production shopper behavior like product views, products added to a cart, and checkouts. Adobe Sensei then uses this information to create and train machine learning models for each website and store view. This unlocks recommendation types like “Customers who viewed this, also viewed…”, which automatically adjusts with shopper behavior over time. Magento and Adobe Sensei do not collect personally identifiable information.

• saas-export — This module syncs catalog data. This type of data provides product information to the Product Recommendations service so it can accurately return product names, pricing, images, URLs, inventory and availability, and other attributes.

  If you prefer, you can install the above modules explicitly using Composer: composer require magento/data-services and composer require magento/saas-export

Add Page Builder support

Product Recommendations for Page Builder is an optional module and is installed separately. To use Product Recommendations with Page Builder, install the module by running the following command:

1
composer require magento/module-page-builder-product-recommendations

By enabling Product Recommendations in Page Builder, you can add an existing, active recommendation unit to any content created in Page Builder, such as pages, blocks, and dynamic blocks.

Add Visual similarity recommendation type

The Visual similarity recommendation type allows you to deploy a recommendation unit to your product detail page that displays products that are visually similar to the product being viewed. This recommendation type is most useful where images and visual aspects of the products are important parts of the shopping experience. Install the Visual similarity recommendation type by running the following command:

1
composer require magento/module-visual-product-recommendations

Update your Product Recommendations installation

Like all of Magento, Product Recommendations uses Composer for installation and updates. To update the magento/product-recommendations module, run the following:

1
composer update magento/product-recommendations --with-dependencies

To update to a major version, such as from 2.0 to 3.0, you must edit your project’s root composer.json file. (See the release notes for information about the latest version.) For example, let’s open the main composer.json file and search for the magento/product-recommendations module:

1
2
3
4
5
"require": {
    ...
    "magento/product-recommendations": "^2.0",
    ...
}

Let’s bump the major version from 2.0 to 3.0:

1
2
3
4
5
"require": {
    ...
    "magento/product-recommendations": "^3.0",
    ...
}

Save the composer.json file and run:

1
composer update magento/product-recommendations --with-dependencies

Uninstall Product Recommendations

If needed, you can uninstall the product-recommendations module.

Configure Product Recommendations

After you install the magento/product-recommendations module, you must configure the module by specifying the API Key and selecting a SaaS Environment.

To ensure catalog export is running correctly, confirm that the cron jobs and the indexers are running and the Product Feed indexer is set to Update by Schedule.

When you successfully link to Magento Services through the API key and specify the SaaS Environment, the catalog sync initiates and behavioral data collection begins on your storefront.