Define the service to recover
List the application, identity services, network routes, databases, interfaces, printing and reporting dependencies. A restored database is not the same as a working clinical service.
Agree recovery objectives
Document the maximum acceptable data loss and the target time to restore service. Measure the drill against those objectives rather than changing the definition of success after the event.
Time the full recovery
Capture the start, data availability, application login, essential workflow completion and background processing milestones. Report them separately. Do not present an early technical milestone as full operational recovery.
Validate with operational users
Ask representative users to complete agreed test journeys. Check recent records, orders, results, charges and interface queues. Use approved test data and retain evidence of reconciliation.
Practise the return to normal
Plan failback, communication, data reconciliation and the recovery of queued transactions. Every failed check needs an owner, a due date and a retest decision.
A checklist to use with your team
- Recovery objectives approved
- Dependencies included
- Operational workflows tested
- Data and interface reconciliation completed
- Failback rehearsed
- Findings assigned and retested
Continue with healthcare digital transformation or AI for healthcare.