Use the Command Center

Before you begin

Sign in and confirm that your team membership and incident permissions are current. Counts are authorization-scoped: two users can legitimately see different results. Choose a known incident and service if you need to validate the dashboard against source records.

Open the feature

Select Dashboard in the main navigation or open /. Read the system-status indicator first, then confirm the displayed time range and active filters before interpreting any number.

Configure the view

Select the required preset or custom time range.
Narrow the population by service, assignee, urgency, or text search when the controls are available to your role.
Clear filters before treating a number as workspace-wide.
Follow a metric card or queue item into its incident list and preserve the resulting query parameters when sharing the view.

The headline cards distinguish incidents in the selected range from current state. Active combines triggered and acknowledged incidents; muted state can include snoozed and suppressed incidents. Resolved is measured for the selected window. Unassigned identifies incidents without an assignee.

How the Command Center works

System status summarizes the current operational condition.
Ops Pulse groups the most immediate responder signals.
My Queue prioritizes incidents assigned to the signed-in responder.
Critical Focus surfaces high-impact work requiring attention.
Services at Risk groups services affected by active incident pressure.
Smart Insights provides computed observations; validate an insight against the underlying incidents before using it in a decision or report.
On-call context shows current coverage based on schedules visible to you.
SLA Alerts shows approaching or breached acknowledgement/resolution targets and links the responder to the affected work.
Quick Actions opens the Create Incident modal or navigates to Analytics and Services. Creating an incident is a mutation; verify the form before submit.
Incident History Heatmap visualizes incident volume across the displayed historical buckets. Treat color intensity as a comparative signal and inspect the underlying incidents before drawing a conclusion.
Latest incidents is a read-only recent-incident table scoped by the active dashboard filters and time window. Open a row to respond on the incident page.
Performance summarizes MTTA, MTTR, acknowledgement-SLA compliance, and resolution-SLA compliance for the dashboard scope. Empty/unavailable data is different from a measured zero.
Team Load summarizes operational load by responder/team using the current analytics snapshot; use it to investigate distribution, not as a standalone performance assessment.
Who's On-Call lists current visible shifts. Validate coverage on the owning schedule before relying on it for a handoff.

Realtime updates can refresh current incident metrics after incident events. When population filters are active, the client requests filtered metrics so a live workspace-wide count does not silently replace the scoped count. The view also exposes an “as of”/data-state context where the metric contract supplies one. Retention can clip historical ranges; a clipped result is not a complete all-time record.

Verify the result

Open a headline count and compare the returned incident list with its label, selected range, and filters.
Change one controlled test incident and confirm the current-state card updates after the realtime event or a manual refresh.
Compare My Queue with the same assignee filter on the Incidents page.
Open an SLA warning and verify the incident deadline and state.
Compare the heatmap and Latest incidents with the same window on Incidents.
Compare Performance and Team Load with Analytics using equivalent scope.
Follow an on-call entry to its schedule and confirm current coverage.
If your role permits export, export the selected view and confirm its scope matches the visible filters rather than assuming it is workspace-wide.

Change or undo the view

Filters and ranges change the view, not incident records. Clear each filter or return to the default range to restore the baseline. An incident action reached through a dashboard link is a separate mutation and must be reversed, where the incident workflow permits, from the incident itself.

Troubleshooting

A count differs from another user's count: compare roles, team membership, service access, filters, range, timezone, and retention clipping.
A count looks stale: wait for the live update, use refresh, then verify the source incident. A degraded live channel must not be treated as proof that no incident changed.
A panel is empty: clear filters and confirm relevant seeded or production records exist and are visible to the user.
Export is unavailable: the current role or visibility scope may not allow export. Ask an administrator to review permissions; do not broaden access only to obtain a report.
An SLA number is unexpected: compare the incident's immutable selected SLA targets and lifecycle timestamps with the current policy; later policy changes do not rewrite an existing incident's contract.