There is a tier of disaster recovery beyond backups, and most businesses first hear about it from a questionnaire or a salesperson rather than a plain explanation. Backups restore your data; Disaster Recovery as a Service (DRaaS) keeps a continuously updated copy of your entire systems (servers, applications, configuration, data) standing by in a provider’s cloud, ready to be switched on when yours fail. The difference in outcome: instead of rebuilding and restoring for a day or three, your business fails over and keeps working, often within the hour.
That capability has a cost, and the honest question is never “is DRaaS good?” (it is) but “is our downtime expensive enough to justify it?” This guide explains how it works, what it costs, and how to answer the fit question without a salesperson’s help.
How DRaaS actually works
Three mechanisms make it up:
Replication. Software continuously copies your production systems (whole virtual machines, not just files) to the provider’s cloud platform, typically every few minutes to hours depending on the tier. This is what keeps the standby copy current, and it is the RPO dial: tighter replication, smaller data-loss window.
Standby infrastructure. The replicas sit dormant on the provider’s platform, consuming cheap storage rather than expensive running compute. This dormancy is the economic trick that makes DRaaS affordable: you pay standby rates for capacity you only fully consume during a disaster.
Failover and failback. When disaster strikes (server room flood, ransomware, hardware death), the replicas are powered up in the cloud and your staff connect to them, working from the copy while the original is rebuilt. Once production is repaired, changes sync back (“failback”) and normal service resumes. Good arrangements script and rehearse both directions, because an untested failover is a hope with a subscription fee, and failback is where amateur setups get stuck.
The net effect on your recovery numbers: RTOs drop from days (restore-and-rebuild) to minutes-to-hours (switch on the standby), with RPOs of minutes on tighter tiers.
DRaaS vs backup: not a replacement
A point that saves expensive confusion: DRaaS complements backup, it does not replace it. Replication faithfully copies whatever happens to production, including the ransomware encryption, within minutes. Point-in-time recovery, long retention and the immutable copy that survives an attacker with admin credentials remain backup’s job, per the 3-2-1-1-0 discipline. Mature setups run both: DRaaS for continuity of *service*, backup for integrity of *data*. (Better DRaaS platforms keep multiple restore points precisely to soften the replicated-ransomware problem, but they are a complement to immutable backup, not a substitute.)
Who genuinely needs it
Run the arithmetic from your downtime cost per hour, because DRaaS is precisely priced insurance against exactly that number:
Strong fit: businesses whose hourly downtime cost is large and legible: manufacturers with stopped lines, e-commerce with a dead checkout, practices with statutory deadlines, care providers with clinical stakes, and any business whose DR plan honestly concluded the RTO must be measured in hours while their current arrangement delivers days.
Poor fit: businesses that could trade tolerably for a day or two on workarounds while systems restore from good backups. For them, the money buys more protection spent on tighter backup, tested restores and the security layers that make disasters rarer. Cloud-native businesses (everything in Microsoft 365 and SaaS) mostly do not need DRaaS at all; their continuity problem is M365 backup and connectivity, not server failover.
The tell that you are in the first group: you have read a recovery-time estimate for your current setup and felt your stomach drop.
What it costs
DRaaS pricing is built per protected server/VM plus replicated storage, with UK SME arrangements typically landing in the low-to-mid hundreds of pounds per month for a handful of protected servers: standby storage is cheap, and the full compute bill only arrives during an actual failover (and then only for its duration). Against the alternative capable of similar RTOs (a second site with duplicate hardware) it is a fraction of the cost, which is why the model exists.
Two contract points to check: test entitlement (how many rehearsal failovers per year are included, because you will want at least one, and a provider discouraging tests is telling you something) and declared-disaster costs (what running in failover actually bills per day, so a two-week recovery is a known number, not a second disaster).
Frequently asked questions
What is DRaaS in simple terms?
A continuously updated standby copy of your servers held in a provider’s cloud, switched on when yours fail so the business keeps operating while the originals are repaired. Recovery time drops from days to minutes-or-hours.
What’s the difference between DRaaS and cloud backup?
Backup stores recoverable copies of data for restoring later; DRaaS maintains runnable copies of whole systems for immediate failover. Backup protects data integrity (including against ransomware, via immutable copies); DRaaS protects service continuity. Serious setups use both.
Does DRaaS protect against ransomware?
Partially: failover to a pre-infection restore point can shorten recovery dramatically, but replication also copies encryption quickly, so DRaaS alone is not the ransomware answer. Immutable backup plus DRaaS covers both halves; see our ransomware guide.
How often should failover be tested?
At least annually as a full rehearsal, with the results timed and documented; the test entitlement in your contract exists to be used. An untested DRaaS arrangement shares the defect of an untested backup: it is a belief.
What does DRaaS cost for a small business?
Typically low-to-mid hundreds of pounds monthly for a few protected servers, scaling with server count and storage, plus usage billing during actual failovers. Priced against your hourly downtime cost, the fit question usually answers itself in one direction or the other quite loudly.
Can DRaaS work alongside private cloud hosting?
Naturally: replicating private-cloud-hosted systems to a second location is a common and clean architecture, and for businesses already hosting with us the DR layer attaches without new plumbing.
Price it against your worst Wednesday
The decision needs two numbers: what an hour of downtime costs you, and what your current tested recovery time actually is. Our free IT health check establishes both, and if the gap says DRaaS, we will design and price it; if it says better backups, we will say that instead, because selling failover to a business that needs restore-testing helps nobody. Get in touch and bring your worst-case afternoon with you. Our disaster recovery service covers the full spectrum either way.
