Incident Follow-Up Tracking

Action items shipped on time. The tracking that prevents 'it'll happen' theater.

Dashboard

The dashboard is the single visible surface for postmortem action-item health. Per-team open, completed, and overdue counts surface where the work is and where it has stalled.

• Per-team open, completed, overdue. Three counts per team. Open shows current load; overdue shows the failure mode.
• Visible to leadership. Org-wide visibility replaces “we’ll get to it” with honest accountability.
• Named owner per action. Single person responsible. Collective ownership rots; named ownership ships.
• Explicit deadline per action. Due date in the tracker. Overdue tracking only works with deadlines that exist.

Healthy rate

The on-time completion rate is the discipline benchmark. Below the floor, action items are theatre rather than work; the postmortem ritual continues but the system stops improving.

• 70 percent on-time. Healthy delivery rate per team. Reasonable bar for committed work.
• Below 50 percent. Not-real signal. Action items are not being treated as real engineering work.
• Trend chart per team. Quarterly trajectory catches degrading delivery early.
• Capability profile per team. Patterns in completion reveal systemic constraints. Some failure modes are scope, not effort.

Escalate stale

Stale action items need explicit escalation. Monthly review surfaces them; repeated patterns trigger leadership conversations about staffing, scope, or priority.

• Monthly review. Overdue audit per month. Surfaces stale items before they become invisible.
• Repeated overdues. Recurring pattern per team triggers conversations about capacity and priority.
• Disposition per action. Close, reschedule, or escalate decision per overdue item. Catches zombie tickets.
• Capacity check per team. Quarterly workload review supports honest scoping next time around.