Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Monitor system refresh performance by using the Performance page. This page provides a centralized view of your Customer Insights - Data environment's refresh health. Use it to quickly assess the overall system status, identify slow or failing pipeline stages, and review refresh history over time.
To open the Performance page, select Performance in the left navigation pane. The page contains the following sections:
- Page header
- Overall health
- Overview tab
- Timeline tab
- Breakdown by stage tab
Page header
The page header at the top of the page shows the time, outcome, and total duration of the most recent refresh. For example: Last refresh: 4/23, 02:15 AM — Successful (44m 40s).
Overall health summary
The Overall health bar summarizes the outcome of all tasks in the most recent refresh. It shows task counts for each status. For definitions of each status value, go to Status definitions.
Below the health summary, each stage shows its task count, duration, and status from the most recent refresh.
Each stage represents the major pipeline stages that run as part of a system refresh:
- Data sources
- Unification
- Enrichments
- Segments
- Measures
- Models
- Sync to Dataverse
- Export destinations
Select a stage to filter the task list on the Overview tab to that stage only.
Overview tab
The Overview tab is the default view and contains two panels: stage cards and task list.
Stage cards
Each stage has a card that shows:
- Stage name and overall status.
- Count of tasks broken down by status (for example, 2 Successful, 1 Completed with warnings).
- A proportional progress bar.
- Total duration for the stage.
Select a card to open the detail view for that stage.
Task list
The task list shows all tasks across all stages for the most recent refresh. To show only tasks for a specific status, select the status filter. Or use the Task search box to filter by task name.
The list includes the following columns:
| Column | Description |
|---|---|
| Task | Task name and pipeline stage in parentheses. |
| Status | Outcome of the most recent run. See Status definitions. |
| Duration | How long the task took to complete. |
| Last updated | Date and time the task last ran. |
Select a task row to view its detail, including logs and error information.
Timeline tab
The Timeline tab shows a Refresh duration trend, which is a historical log of completed system refreshes.
Each row represents one full system refresh and includes:
| Column | Description |
|---|---|
| Time range | Start and end time of the refresh |
| Status | Overall outcome (for example, Successful, Canceled) |
| Duration | Total elapsed time |
| Scope | The deepest pipeline stage that ran (for example, Merge, Match, Data sources) |
| Mode | Incremental or Full |
| Change | Percentage change in duration compared to the previous refresh of the same scope |
| Changes | Details of tasks that were added or removed from this refresh |
Rows marked with a configuration change indicate that one or more tasks were added or removed in that refresh.
You can:
- Change the time window for the refresh history: select 3 days, 7 days, or 30 days.
- View a per-stage Gantt timeline for a specific refresh: select a row to expand an inline timeline that shows each pipeline stage running in parallel, with the individual task results within each stage.
Breakdown by stage tab
The Breakdown by stage tab shows the same historical refresh list as the Timeline tab, and adds a proportional time breakdown chart for each refresh.
To filter by stage, select one of the stages at the top of the tab: Data sources, Unification, Segments, Measures, or Sync to Dataverse.
Each expanded refresh row shows a bar chart of how the total duration was distributed across stages (for example, unification: 29m 7s, dataverseHydration: 10s). Use this view to identify which stage is the bottleneck in a given refresh.