An Azure data analytics service for real-time analysis on large volumes of data streaming from sources including applications, websites, and internet of things devices.
There’s no built-in feature in Azure Data Explorer Web UI that directly reports “time spent” per user on dashboards or panels. However, the platform provides the building blocks to approximate engagement using Kusto logs and user analytics plugins.
A practical approach is:
- Log dashboard access events
- When dashboards are viewed via the Azure Data Explorer web UI, the underlying activity is KQL query execution against the cluster.
- Use a telemetry/logging mechanism (for example, Application Insights or a custom log table in ADX) to capture at least:
- User identifier
- Dashboard/panel identifier
- Timestamp of each view or interaction event
- Model sessions and engagement windows in KQL
- Once events are logged into a table in ADX, use Kusto’s user analytics plugins to derive engagement metrics.
- For “how often” and “how many users” use dashboards (DAU/WAU/MAU-style engagement), use the
activity_engagementplugin:
This calculates activity engagement ratios over sliding time windows and can be visualized in Azure Data Explorer dashboards.Events | evaluate activity_engagement( "UserId", // user identifier column "Timestamp", // event time column 1d, // inner window (e.g., daily activity) 7d // outer window (e.g., weekly activity) )
- Approximate “time spent” from event sequences
- To estimate time spent per dashboard:
- Treat each view/interaction as part of a session.
- For each user and dashboard, sort events by time and compute the time difference to the next event or a max cap (for example, 10–15 minutes of inactivity ends the session).
- Sum these durations per user and dashboard.
- This logic is implemented with standard KQL (sessionization) on top of the logged events; the
session_countplugin (listed in the user analytics plugins overview) can help count sessions, and custom KQL can extend that to approximate duration.
- To estimate time spent per dashboard:
- Visualize engagement in ADX dashboards
- Use the Azure Data Explorer web UI query page to build KQL queries that:
- Aggregate estimated time spent per user, per dashboard, per day/week.
- Compute engagement ratios (DAU/WAU/MAU) with
activity_engagement.
- Export these queries as tiles into Azure Data Explorer dashboards for ongoing monitoring.
- Use the Azure Data Explorer web UI query page to build KQL queries that:
This approach uses ADX as both the storage and analytics engine for engagement telemetry: events are ingested into a table, KQL and user analytics plugins derive engagement metrics, and Azure Data Explorer dashboards visualize the results.
References: