Key takeaways
- RPO measures acceptable data loss; RTO measures acceptable downtime
- Set targets per system, not one number for everything
- Backup frequency drives RPO; the restore method drives RTO
- Time real restores to prove the targets are met
RPO: how much data you can lose
RPO looks backward from the moment of failure to the last good backup. If backups run once a night and a server fails late in the afternoon, everything entered since that backup is gone. That gap is your real RPO.
A shorter RPO means more frequent backups or replication, which costs more. The right target depends on what the lost work is worth: re-entering a day of invoices is an annoyance, while losing a day of patient charts or client trades is a serious problem.
RTO: how long you can be down
RTO looks forward from the failure to the moment people can work again. It includes noticing the problem, deciding to restore, getting hardware or cloud capacity ready, restoring the data and checking that it works.
Restoring a large server from cloud storage alone can take far longer than people expect. A local copy, a spare device or the ability to run a system in the cloud while the original is rebuilt is what shortens RTO.
How to set your targets
List your systems
Email, file shares, line-of-business apps, accounting, phones and anything else people use daily.
Ask what an outage costs
For each one, what happens after a morning down, a day down or a week down? Who is affected and what stops?
Rank them
Group systems into tiers. The few that stop the business get the tightest targets, and the rest can wait their turn.
Match the method
Choose how often to back up for the RPO and how to restore for the RTO, then weigh the result against budget.
Test and adjust
Run a restore or failover test and time it. If the result misses the target, change the method or the target.
A typical example
A typical 25-person professional services firm puts its practice management system and file server in the top tier, backed up through the workday with a local copy for a fast restore. Email in Microsoft 365 sits in a middle tier. Archived projects go in a lower tier, where a slower restore from the cloud is acceptable.
Writing the tiers down turns a vague worry into decisions the owner can sign off on, and gives the next test something to measure against.
How we help
Our backup and disaster recovery add-on starts with a defined RPO and RTO for your systems, then builds the local and offsite backup strategy around them. Quarterly or annual failover testing checks the plan against those targets, and quarterly backup summary reports show how it is going. Disaster recovery and continuity planning are part of the same add-on.
Sources
Information on this page is general and is not legal or compliance advice.