Outages, ransomware and regional cloud incidents are no longer rare events. Most organisations have backups. Far fewer can say with confidence how long it would take to restore critical services, in what order and who would make the decisions.
Start with business impact
Identify critical services with business owners and agree two numbers for each: the recovery time objective, how long the service can be unavailable, and the recovery point objective, how much data loss is acceptable. These numbers drive every design decision and its cost.
Design for the threats you face
- Hardware and site failure. Replication to a secondary site or cloud region.
- Ransomware. Immutable backups, isolated recovery environments and clean recovery procedures.
- Human error. Point in time restores and change controls.
- Cloud regional incidents. Cross region designs for the most critical workloads.
Test, then test again
Schedule restore tests for critical systems at least quarterly and a full recovery exercise at least annually. Record the actual recovery times, compare them with objectives and fix the gaps.
Make it operational
Recovery runbooks, contact lists and decision rights should live where people can reach them during an incident, and be reviewed whenever systems change.
Our Backup and Disaster Recovery services design, test and operate recovery capability for on premises and cloud estates.
