Incident and Program Leader playbook

The incident or program leader manages the operating system around the technical path. They make sure every important question, action, approval, communication, and deadline has an owner. They do not replace the technical decision maker.

Use the tunnel and dossier to identify unowned branches, overdue evidence, blocked approvals, and cross-team dependencies. A status such as "engineering investigating" is not sufficient. Name the evidence being produced and the next decision it enables.

Time-box a checkpoint according to the cost of waiting and the time required to produce reliable evidence. Short checkpoints that only repeat "still investigating" create noise. Long checkpoints without intermediate proof create risk.

Align internal and external statements with the same evidence state. Preserve uncertainty where uncertainty exists. Record the next commitment and who approves the message. Avoid parallel narratives that cause the customer, vendor, and engineering teams to work from different assumptions.

Checkpoint plan, dependency map, decision log, communication cadence, escalation status, and closure readiness summary.