A successful backup job proves that bytes moved somewhere. It does not prove that the business can recover. Recovery depends on data integrity, infrastructure, identity, networking, secrets, dependencies, people, and a sequence of decisions made under pressure.
That is why a disaster recovery plan is best treated as a falsifiable theory. Exercises exist to find the conditions under which the theory fails.
Define recovery in business terms
Recovery objectives should describe the capabilities the organization needs, not merely the infrastructure it wants restarted. Which workflows must return first? What data loss is tolerable? Who can declare a disaster, and what information do they need?
Map those answers to measurable recovery time and recovery point objectives, then trace every dependency required to meet them.
Practice the uncomfortable parts
A good exercise includes damaged assumptions: the usual administrator is unavailable, a credential has expired, documentation is stale, or the primary cloud control plane is impaired. The aim is not theater. It is to discover fragile knowledge and hidden coupling while the stakes are still low.