Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
Recovery time objective is how long a service may stay down; recovery point objective is how much data you can afford to lose. Together they set what your backup design must achieve.

What Are RTO and RPO?
Recovery time objective (RTO) is the maximum time a service may remain unavailable before the consequences become unacceptable. Recovery point objective (RPO) is the maximum amount of data, measured in time, that the organisation is willing to lose. One is answered with a clock, the other with a window.
An order system with a four-hour RTO and a 15-minute RPO must be back within four hours of failing, having lost no more than the last quarter of an hour of transactions. Those two numbers determine almost everything about the architecture behind it.
Place the moment of failure on a timeline. RPO measures backwards from it to the last recoverable copy of data; RTO measures forwards to the point where the service is usable again. They are independent: you can have continuous replication with a near-zero RPO and still take two days to restore service because nobody documented how to rebuild the application tier.
Two related terms often sit alongside them. Maximum tolerable period of disruption is the absolute outer limit beyond which the organisation suffers irreversible harm, and RTO is deliberately set inside it. Work recovery time covers the validation, reconciliation, and catching up that follow a technical restore, and teams that ignore it consistently promise recovery times they cannot meet.
RTO and RPO are business decisions, not technical preferences. They come out of a business impact analysis, which asks each service owner what happens hour by hour when the service is unavailable: lost revenue, regulatory breach, safety consequences, contractual penalties, reputational damage. The analysis is what stops every department declaring that its system is critical.
Most organisations then group services into tiers rather than setting unique objectives for each:
| Tier | Example | RTO | RPO | Typical design |
|---|---|---|---|---|
| 1 | Payment processing | Under 1 hour | Near zero | Active-active across regions, synchronous replication |
| 2 | Customer portal | 4 hours | 15 minutes | Warm standby, asynchronous replication |
| 3 | Internal reporting | 24 hours | 24 hours | Nightly backup, restore to cloud on demand |
| 4 | Archive systems | 5 days | 7 days | Weekly backup to cold storage |
Cost rises steeply as the numbers fall. Going from a 24-hour RPO to 15 minutes usually means replacing nightly backups with log shipping or replication; going to near zero means running a second live environment. That is why tiering, and a considered backup strategy, matters more than applying the strictest figure everywhere.
Traditional disaster recovery planning assumed a failure with a clean starting point: a flood, a hardware failure, a lost data centre. Ransomware breaks that assumption in three ways. The most recent backup may be encrypted or deleted, because attackers target backup infrastructure first. The most recent clean backup may predate the intrusion by weeks, destroying your stated RPO. And recovery cannot begin until forensics establish how the attacker got in, or you restore straight back into a compromised environment.
The practical responses are well established: immutable or object-locked backups that cannot be deleted inside their retention period, at least one offline or logically isolated copy as the NCSC recommends, separate credentials for backup systems, and a rehearsed plan for rebuilding identity infrastructure. Restoring Active Directory from scratch is frequently the longest task in a real recovery, and it rarely appears in the plan.
An RTO is a claim until you have tested it. ISO 22301, the business continuity management standard, requires exercises and reviews for exactly that reason, and the honest version of the test restores a full service to a separate environment and times it, including the work recovery phase. Partial file restores prove very little.
Run a tabletop exercise alongside the technical test so that decision-making is rehearsed too, and feed the measured times back into the plan. If the tested recovery takes nine hours against a four-hour objective, you have two choices: invest in the gap or change the objective to something you can defend to a regulator. Publishing an RTO you have never achieved is the worst of both.
If you want these objectives tested rather than assumed, our emergency action plan services rehearse recovery against the times you have committed to.
RTO measures downtime: how long a service may be unavailable before the impact becomes unacceptable. RPO measures data loss: how far back in time the last usable copy of data may be. A system can meet one and fail the other, for example when replication keeps data current but the recovery procedure takes a full day to execute.
Through a business impact analysis rather than a conversation with the technical team. Ask the service owner what happens at one hour, four hours, one day, and one week of unavailability, in terms of revenue, regulatory exposure, safety, and contractual penalties. The point at which consequences become severe sets the maximum tolerable period, and the RTO is set comfortably inside it.
Not by itself. Continuous replication faithfully copies encrypted files to the secondary site, and attackers commonly delete or encrypt backups before triggering the payload. Protection comes from immutable retention, an offline or logically isolated copy, separate credentials for backup systems, and the ability to restore to a point before the intrusion began rather than before the encryption ran.
At least annually for every tier 1 and tier 2 service, and after any significant architecture change. The test should restore a complete service to a separate environment and record the actual elapsed time, including validation and reconciliation. Testing only file-level restores gives false confidence, because the slow parts of a real recovery are rebuilding dependencies and identity systems.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.