Recommendation dashboard

This is for:

Business Analyst System Administrator

The Recommendation dashboard on the Advanced Reports (platform-ca | platform-eu | platform-au) page provides an in-depth view of the top-performing recommendations. As in the Overview dashboard, you can view basic metrics such as clickthrough rates, click ranks, and more. However, the Recommendation dashboard provides an emphasis on recommendation attributions.

The Recommendation dashboard includes the following metric charts:

What is a session?

Although sessions aren’t specifically listed in the Recommendation dashboard, they’re often used to calculate the metrics listed there.

At Coveo, a session starts with the first event a visitor generates and ends after 30 minutes of inactivity. For a given visitor, if a new event is generated after 30 minutes of inactivity, it results in a new session with a new visit ID.

To be considered Coveo-related, the session must contain search, product listing, or recommendation actions.

What is a recommendation event?

In this article, the term recommendation event is often used in metric definitions.

The definition of a recommendation event varies based on whether your Coveo for Commerce implementation targets the Commerce API or the Search API. Expand the following sections to learn about recommendation events based on the Coveo API your implementation targets:

Commerce API

A recommendation event is triggered when a request is made to the recommendations endpoint of the Commerce API.

Search API

In practice, a recommendation event can be any of the following:

What is recommendation attribution?

A click action on a recommended product | Coveo for Commerce Coveo attributes a product in a transaction to the Recommendation product discovery solution if the visitor's last registered click action before purchasing the product was selecting it from a recommendation slot.

For more details about how attribution is tracked for the Recommendation solution, see About recommendation attribution.

The Recommendation dashboard shows how Coveo-powered product recommendations have contributed to your key commerce metrics.

Item revenue from recommendations chart

The Item revenue from recommendations chart displays the total revenue influenced by Coveo recommendations that occurred within the chosen filters.

The line graph delineates the revenue variance over a specific period of time.

How is the item revenue from recommendations calculated?

Expand the following sections to learn how the item revenue from recommendations is calculated, based on whether you’re tracking events using the Coveo Event protocol or the Coveo UA protocol:

Event protocol

Item revenue from recommendations is calculated by adding up the revenue (from the transaction property) from every product sold in each purchase event that was attributed to recommendation events.

Coveo UA

Item revenue from recommendations is calculated by adding up the revenue from every product sold in each purchase event that was attributed to recommendation events.

Note

Revenue is reported in USD and calculated as either net or gross revenue:

  • Net revenue excludes taxes, shipping costs, and discounts.

  • Gross revenue includes taxes, shipping costs, and discounts.

The Top selling products metric displays net revenue figures. All other reported revenue metrics use gross revenue, unless otherwise stated.

See About revenue for more information on how the revenue metrics are calculated.

Transactions from recommendations chart

Expand the following sections to learn about the Transactions from recommendations chart based on the tracking method you use:

Event protocol

The Transactions from recommendations chart displays the total number of purchase events influenced by Coveo recommendations that occurred within the chosen filters.

The line graph delineates the variance of the purchase events over a specific period of time.

Coveo UA

The Transactions from recommendations chart displays the total number of purchase events influenced by Coveo recommendations that occurred within the chosen filters.

The line graph delineates the variance of the purchase events over a specific period of time.

How are transactions from recommendations calculated?

Expand the following sections to learn how transactions from recommendations are calculated, based on the tracking method you use:

Event protocol

The total number of purchase events attributed to recommendation events.

See Capture purchase events for more information on how purchase events are logged and measured.

Coveo UA

The total number of commerce events that contain the purchase action and are attributed to recommendation events.

See Capture a purchase event for more information on how purchase events are logged and measured.

Conversion rate from recommendations chart

The Conversion rate from recommendations chart displays the number of sessions that involved Coveo recommendations and in which at least one transaction occurred over the total number of sessions that involved Coveo recommendations, within the chosen filters.

The line graph delineates the variance of the rate over a specific period of time.

How is the conversion rate from recommendations calculated?

Number of sessions involving Coveo recommendations with at least one transaction, divided by the total number of sessions in which at least one recommendation event occurred, multiplied by 100.

Average order value influenced by recommendations chart

The Average order value influenced by recommendations chart displays the average revenue per transaction influenced by Coveo recommendations that occurred within the chosen filters.

The line graph delineates the variance of the average over a specific period of time.

How is the average order value influenced by chart-name calculated?

The sum of the total revenue attributed to recommendation events, divided by the total number of transactions attributed to recommendation events.

Average clickthrough rate chart

The Average clickthrough rate chart displays the ratio between the number of recommendation events that were followed by at least one related click event, out of the total number of recommendation events, within the chosen filters.

The line graph delineates the variance of the average clickthrough rate over a specific period of time.

How is the average clickthrough rate calculated?

The number of recommendation events that were followed by at least one click event (with a matching unique identifier value), divided by the total number of recommendation events.

Recommendation components chart

The Recommendation component chart displays a list of the recommendation slots with the most interactions, as well as the revenue generated by each component, within the chosen filters.

How are recommendation components calculated?

For each recommendation component, the chart displays the following metrics:

Total recommendation events chart

The Total recommendation events chart displays the total number of recommendation events, within the chosen filters.

The line graph delineates the variance of the total over a specific period of time.

Top selling products chart

A list of the top ten products or product groups in recommendation slots that generated the most revenue, within the chosen filters.

How are top selling products calculated?

The products or product groups contained in click events that followed a recommendation event to which a purchase was attributed and that generated the most revenue.

Discrepancies between reports

Figures shown for metrics in the Commerce Advanced Reports may differ from those for the same metrics in the Reports (platform-ca | platform-eu | platform-au) page of the Coveo Administration Console and Coveo Merchandising Hub (CMH) reports.

These discrepancies typically occur because different reports use distinct filters and processing rules when calculating figures for the same metric.

Here are the most common reasons for differences between reports:

  • Inclusion of searchBoxAsYouType events:

    By default, reports from the Reports (platform-ca | platform-eu | platform-au) page include search events where the event cause is searchBoxAsYouType. This can result in higher query figures compared to the Commerce Advanced Reports or CMH reports, which exclude these events by default.

  • Time zone differences:

    CMH reports use Coordinated Universal Time (UTC), while the reports of the Reports page and Commerce Advanced Reports use the user’s browser time zone. This difference can lead to a few hours of offset in event timestamps, which may affect metric calculations.

  • Case sensitivity in query metrics:

    For query-related metrics, Commerce Advanced Reports and CMH reports are case-sensitive, while the reports on the Reports page are case-insensitive. This means that identical queries with different casing are treated as separate queries in Commerce Advanced Reports and CMH reports, but are counted as the same query in the reports on the Reports page.

    For example, the queries coveo and Coveo are counted as separate queries in Commerce Advanced Reports and CMH reports, but as the same query in the reports on the Reports page.