AI Search analytics dashboard
Summarize
Summary of AI Search Analytics Dashboard
The AI Search analytics dashboard provides key performance metrics, trends, and reports for AI Search usage, enabling analysts to monitor and improve search functionality. Users can apply interactive filters to analyze performance for specific search applications and select desired time frames.
Show less
Key Features
- Accessing the Dashboard: Navigate to All > Usage Insights, select the application's Launch Dashboard link, and then select the Search Analytics icon or link.
- Required Roles: Users must have the aisadmin role and one of the following: analyticsadmin, analyticsviewer, portalanalyticsadmin, portalanalyticsviewer, webanalyticsadmin, or webanalyticsviewer.
- Interactive Filters: Filters include selecting a search application and setting a date range for analytics metrics. The dashboard displays data from the last 30 days by default, with a maximum retention of 180 days.
- Metrics and Reports: The dashboard offers insights on unique search users, total queries, average response times, and self-solved rates, among others.
- Search Queries Section: Users can view submitted search queries, along with their click positions and results, and identify queries that yielded no results or low-quality responses.
Key Outcomes
By utilizing the AI Search analytics dashboard, organizations can:
- Monitor and enhance search performance for specific applications.
- Identify trends and user needs to improve search experiences.
- Analyze common queries that result in poor outcomes, allowing for targeted improvements.
- Utilize data-driven insights to optimize overall search functionality across various platforms.
The AI Search analytics dashboard displays key performance metrics, trends, and reports for AI Search usage. Analysts can apply interactive filters to view performance analytics for individual search applications and to select the time frame to analyze.
In the Queries section of the dashboard, you can view the query lists on their own page by selecting View all. To return to the main dashboard from the sub-page, select AI Search Analytics.
You can improve the visual experience by using the dark theme. For more information, see Working with themes in Next Experience.
Required ServiceNow AI Platform roles
- analytics_admin
- analytics_viewer
- portal_analytics_admin
- portal_analytics_viewer
- web_analytics_admin
- web_analytics_viewer
Use cases
| User | Dashboard use |
|---|---|
Any of the following:
|
|
Interactive filters
| Filter | Description |
|---|---|
| Search application | Specifies the search application configuration for which you want to see search analytics metrics, trends, and reports. You can search for a search application configuration or select its name in the list.
Select Apply to make your filter selection take effect. Note: You can only select a single search application configuration in this filter. The filter list includes search application configurations that are configured to use AI Search and that are used in any of the following contexts:
The dashboard reports metrics for all search queries in the selected search application configuration that satisfy your Date range constraint. These may include user searches submitted directly to a search application that uses the configuration and searches submitted through interaction with other ServiceNow AI Platform features, such as Virtual Agent. |
| Date range | Specifies the range of dates for which you want to see search analytics metrics, trends, and reports. By default, the dashboard displays data from the last 30 days. You can choose one of the preset date ranges
or set your own start and end dates. Select Apply to make your filter selection take effect. The dashboard retains data from the last 180 days. If you select a date from outside of this range, the system displays an error message and the dashboard reverts to the last valid date range selection. If you haven't selected any
valid date range, the dashboard reverts to its default date range. Note: When you select a custom date range, the dashboard may take longer to load than when you select a preset date range. The dashboard reports metrics for all search queries in the selected date range that satisfy your Search application constraint. These may include user searches submitted directly to the search application and searches submitted through interaction with other ServiceNow AI Platform features, such as Virtual Agent. |
Metrics, trends, and reports
| Title | Description |
|---|---|
| Search users | This numeric metric shows the total number of unique users who submitted search queries. The trend describes how the number of unique search users has changed since the start of your selected date range. The chart shows the number of unique search users for dates in your selected date range. Note: Searches performed while impersonating another user are ignored for this metric, trend, and chart. |
| Total queries | This numeric metric shows the total number of search queries submitted. The trend describes how the number of search queries has changed since the start of your selected date range. The chart shows the number of search queries submitted for dates in your selected date range. |
| Searches over time | This chart shows counts for searches in your selected date range. It includes three sets of values:
|
| Average response time | This numeric metric shows the average response time for search queries in milliseconds. The chart shows the distribution of search query response times, broken out into buckets. The trend describes how the average response time for search queries has changed since the start of your selected date range. |
| Average click position | This numeric metric indicates the average position of user-selected search query results. The chart shows the distribution of positions for user selected search query results. The trend describes how the average click position has changed since the start of your selected date range. |
| Self-Solved Rate | This metric indicates the percentage of search queries that produced a search result click (selection). The chart shows the distribution of search queries that did and didn't produce search result clicks. The trend describes how the self-solved rate has changed since the start of your selected date range. |
| Title | Description |
|---|---|
| Search queries | This list shows user-submitted search queries. Each entry displays the following details:
To view entries for these search queries in the Search Events [sys_search_event] table, select View all. |
| Title | Description |
|---|---|
| Top queries | This list shows the most frequently submitted search queries. To view this list on its own sub-page, select View all. To return to the main dashboard from the sub-page, select AI Search Analytics. |
| Queries with no clicks | This list shows the most frequently submitted search queries for which users didn't select any search result. To view this list on its own sub-page, select View all. To return to the main dashboard from the sub-page, select AI Search Analytics. |
| Queries with no results | This list shows the most frequently submitted search queries that produced no results. To view this list on its own sub-page, select View all. To return to the main dashboard from the sub-page, select AI Search Analytics. |
| Title | Description |
|---|---|
| Top clicked results | This list shows the most frequently clicked search results. |