This is the unified documentation for Retail API. This includes Recommendations AI, Retail Search, and the unified Retail console (which is applicable to both Recommendations AI and Retail Search users). To use the new console or Retail Search while they are in the restricted GA phase, submit a form here to contact Cloud sales. If you are using the v1beta version of Recommendations AI, migrate to the GA version: Migrating to the Retail API from beta.

To see documentation for only Recommendations AI and the Recommendations AI-only console, go to the How-to guides for Recommendations AI and the API reference documentation for Recommendations AI.

Measuring performance

Retail provides metrics to help you determine how incorporating the Retail API is affecting your business.

Viewing Retail metrics

You can view summary metrics for your project on the Analytics tab of the Monitoring & Analytics page in Cloud Console. See Summary metrics for metrics definitions.

For serving configuration-specific metrics, go to the Serving Configs page. Click a serving configuration name and select the Analytics tab to see its metrics. See Configuration-specific metrics for metrics definitions.

For details about monitoring Retail API errors and other health-related metrics, see the Retail API Monitoring and troubleshooting documentation.

Metrics definitions

Summary metrics

The tables below provide definitions of the site-wide metrics displayed on the Analytics tab of the Monitoring & Analytics page.

In the Product drop-down list, select Recommendations or Search to view metrics for that type of Retail API product.

Recommendations AI summary metrics

The following metrics are displayed for Recommendations AI.

Select a metric type from the Metric drop-down list to see its graph for the last month.

Metric Description
Total revenue

The total revenue from all recorded purchase events.

This value includes shipping and tax.

Recommender-engaged revenue

The revenue for purchase events that include at least one product that was selected from a recommendation panel.

This value includes shipping and tax and any discount applied.

Average order value (AOV)

The average value of orders from all purchase events.

Total revenue divided by the number of orders.

Recommender-engaged AOV

The average value of orders that include at least one product selected from a recommendation panel.

Recommender-engaged revenue divided by the number of orders with at least one product that was selected from a recommendation panel.

The following metrics are displayed for Retail Search.

Select a metric type from the Metric drop-down list to see its graph for the last month.

Metric Description
Searches

The total number of searches performed site-wide.

Click-through rate

The number of clicks from search results divided by the total number of site-wide searches.

The total number of searches includes paginated queries.

Conversion rate

The number of purchase events occurring from search results divided by the total number of searches.

The total number of searches includes paginated queries.

Add to cart rate

The number of add-to-cart events occurring from search results divided by the total number of searches.

Configuration-specific metrics

You can see metrics for a specific serving configuration on the Serving Configs page. For metric graphs, click a serving configuration name to go to its details page, then select the Analytics tab.

The table below provides definitions for configuration-specific metrics.

Metric Description Details
Click-through rate (CTR) The number of product detail views from a serving configuration's recommendation panel, divided by the total number of `predict` queries for this serving configuration. For example, if the serving configuration points to a "Frequently Bought Together" model, then the CTR would be the number of product detail pages viewed from the shopping cart recommendation panel divided by the number of predict queries on the shopping cart page.
Conversion rate (CVR) The number of add-to-cart events from a serving configuration's recommendation panel divided by the total number of `predict` queries for this serving configuration. Similar to CTR, except that instead of product detail views, conversion rate uses add-to-cart events.
Recommender-engaged revenue The total revenue from the recommendations for this serving configuration. Similar to recommender-engaged revenue for summary metrics, but for this serving configuration only.

To track clicks from a serving configuration, Retail aligns the results in predict and search responses with ingested user events. If a clicked item appears in the responses for the same visitor ID within a one hour time window, the click/purchase is treated as a result of the Retail API.

This process is fully automatic; you do not need to set anything up. However, when you configure your prediction and search requests for the first time, you should confirm that:

  • Visitor IDs in the request are the same as the visitor IDs you used in event ingestions.
  • The timestamp in the response roughly match the timestamp for that event.

When Retail API metrics are compared to the ideal expected result, or ground truth, the values might be lower, but the trends align.

A more direct alternative to the method above is to use attribution tokens. This requires significant instrumentation and is only recommended as an advanced tracking use case.