Tabletop Exercise
A discussion-based walkthrough of a hypothetical incident that tests procedures and decision-making without touching any system.
The cheapest form of resilience testing, and disproportionately effective as a first step. A facilitator presents a scenario; the team talks through what they would do, step by step; the facilitator introduces complications.
Because nothing is actually broken, it can be run in an hour with no risk, no environment preparation and no engineering time beyond attendance — which means it can be run often and can include people who would never join a live exercise: support, legal, communications, and executives who would be making decisions in a real event.
What it reliably finds: nobody knows who declares an incident; the runbook references a system that was decommissioned; the escalation path ends at someone who left; the status page needs credentials one person holds; there is no agreed threshold for notifying customers; and the decision to fail over to another region has no named owner.
None of those require a live system to discover, and all of them add hours to a real incident.
The correct sequence is tabletop first, then simulated failure in a non-production environment, then a controlled production game day. Running a production exercise before the tabletop has been done tends to spend expensive engineering time rediscovering procedural gaps that an hour of conversation would have surfaced.