Your Backups Are Working. But Can You Actually Recover?
Most businesses understand the importance of backing up their data. But how many know whether those backups would actually work when needed?
A backup system might report successful jobs every day, but that doesn’t necessarily mean a failed server could be restored quickly, or that important information could be recovered from several months ago.
At Weald IT, we’ve been providing backup and disaster recovery solutions for many years. One lesson we’ve learned is that having backups and being able to recover are not quite the same thing.
In this article
- Five questions worth asking your IT provider
- Different systems need different backups
- Testing recovery, not just checking backups
- Recovering a complete server
- How quickly do you need to recover?
- How far back can you recover?
- Your backups are your responsibility too
- Backup is only part of business continuity
Five questions worth asking your IT provider
Whether your backups are managed internally or by an external IT company, these five questions are worth asking:
- Are our backups working, and can we see the results?
- When was our last successful recovery test?
- How quickly could our critical systems be restored?
- How much recent data could we potentially lose?
- How far back can we recover, and how many restore points are available?
The answers should be understandable without requiring detailed technical knowledge. The sections below explain why they matter.
Different systems need different backups
Not all backups serve the same purpose.
For businesses using Microsoft 365, it’s important to consider the protection of Exchange mailboxes, OneDrive files and SharePoint data. Recovery might involve restoring an accidentally deleted email, recovering an earlier version of a document or retrieving information following a security incident.
Servers present different challenges. If a server fails, it may be necessary to recover the complete system, including its operating system, applications and data.
At Weald IT, we use solutions including Cove Data Protection and Veeam Backup & Replication, selecting the most appropriate approach according to the client’s infrastructure, requirements and budget.
Testing recovery, not just checking backups
A successful backup report is reassuring, but it doesn’t prove that a system can be recovered.
For smaller servers, we use Cove’s disaster recovery testing facilities. These allow us to start a backed-up server as a virtual machine in an isolated environment and obtain evidence that it can boot successfully.
For larger or more complex environments, we also use Veeam Backup & Replication.
Depending on the client’s requirements, backups and replicas can be maintained locally, with secondary copies or replicas held at Weald’s HQ. These can use our shared storage infrastructure or equipment belonging to the client.
Veeam provides considerable flexibility. For example, a client may have local backups on a NAS or backup server, with an additional off-site backup copy transferred to Weald. Alternatively, virtual machine replicas can be maintained at our premises, or both approaches can be combined.
We can also use immutable backup storage, helping protect recovery points against accidental deletion, malicious alteration and ransomware attacks.
For disaster recovery testing, we can start recovered or replicated virtual machines in an isolated environment, check that they operate correctly and provide evidence of the test.
A backup job completing successfully is one thing; demonstrating that a server can actually start from its backup is another. Starting a virtual machine doesn’t necessarily prove that every application or business process will function correctly, so more comprehensive testing may be appropriate for critical systems.
Recovering a complete server
There are different ways of backing up and recovering servers.
Some solutions use image-based backups, while others, including Cove, use technology that allows complete systems to be reconstructed from backed-up data.
We’ve successfully rebuilt servers using Cove on numerous occasions. Although rebuilding a replacement server can take longer than starting an existing replica, it provides a practical recovery option without requiring dedicated standby infrastructure.
For businesses requiring faster recovery, a standby virtual machine or replica can make a significant difference.
With a suitable replication arrangement, a copy of the server is maintained on alternative infrastructure, potentially at another location. If the original server fails, the replica can be started without having to rebuild the complete system from backup.
This can significantly reduce the Recovery Time Objective (RTO). However, actual recovery time also depends on network connectivity, application dependencies and the arrangements for reconnecting users.
The right approach depends on how quickly the business needs its systems restored and what level of investment is justified. For some organisations, restoring a server over several hours may be acceptable. For others, the ability to start a replicated virtual machine much more quickly may be essential.
How quickly do you need to recover?
Two terms are particularly useful when discussing backup and disaster recovery requirements.
Recovery Time Objective (RTO)
How long can your business reasonably operate without a particular server or system?
A business might tolerate a day’s interruption to a non-critical system, while another may need its main application server restored within an hour or two.
The RTO is a recovery target, not a guarantee.
Where a very short RTO is required, a properly configured standby virtual machine or replica can provide a much faster recovery route than rebuilding a failed server from backup.
Recovery Point Objective (RPO)
How much recent data could your business afford to lose?
For example, if backups run every four hours, a failure could potentially result in losing up to four hours of changes.
Some backup systems, including Cove, can be configured to run every couple of hours where the amount of changing data makes this practical.
Replication can also provide more frequent recovery points, depending on its configuration, although it should not be regarded as a replacement for maintaining separate backups.
Understanding RTO and RPO helps determine whether a straightforward backup solution is sufficient or whether additional replication and disaster recovery arrangements are justified.
How far back can you recover?
Another important consideration is the number and age of available restore points.
Imagine discovering that an important file was corrupted six weeks ago, but nobody noticed until today. If your backups only go back 30 days, the original version may no longer be available.
Modern backup systems can provide extended retention, but this needs to be considered when selecting and configuring a solution.
For example, Cove’s backup technology allows us to retain server backups for extended periods, potentially up to seven years, without necessarily increasing the underlying service cost. This is possible because of the way Cove processes and stores backup data, rather than relying on numerous conventional full backup images.
Veeam also provides flexible retention options, including Grandfather-Father-Son (GFS) policies. These allow businesses to maintain selected weekly, monthly and yearly recovery points alongside their regular backups.
A typical arrangement might include frequent local backups, longer-term GFS retention and a secondary off-site backup copy at Weald’s HQ.
Storage capacity, the rate of data change and the number of retained restore points all influence the appropriate configuration and cost.
More restore points and longer retention can be valuable, but they need to reflect the business’s actual requirements.
Retaining seven years of backups does not automatically mean every file version from every day is available for seven years. The recovery history depends on the configured retention policy.
Your backups are your responsibility too
Although an IT provider may manage a company’s backups, we believe businesses should also have visibility of their own backup arrangements.
Depending on the solution, this could include:
- Access to a backup management console.
- Automated email notifications showing successful or failed backup jobs.
- Regular reports confirming backup status and recovery testing.
- Clear procedures for reporting and resolving backup failures.
A business shouldn’t have to assume everything is working simply because nobody has reported a problem.
We believe backup monitoring should be a shared responsibility, with the IT provider managing the technical processes and the client having sufficient information to understand whether their data is being protected.
Backup is only part of business continuity
Reliable backups are essential, but they’re only one part of a sensible business continuity plan.
Recovery testing, appropriate retention periods, immutable storage, clear responsibilities and realistic recovery expectations all matter.
At Weald IT, we provide backup and disaster recovery solutions for businesses across Sussex, Kent and the surrounding areas, using technologies including Cove and Veeam.
Our approach is to understand what a business needs to protect, how quickly it needs to recover and what level of protection is appropriate.
Already a Weald IT client? If you’re unsure when your servers were last recovery-tested, or whether automated or manual disaster recovery testing would be appropriate, please contact us. We can review your existing backup arrangements and discuss the available options.
Because the real value of a backup isn’t in creating it. It’s in being able to recover when you need it.
Better IT starts with a conversation.
When did you last test your backups?
Whether you’re an existing Weald IT client or looking for advice on backup and disaster recovery, we’re happy to review your arrangements and discuss the options available, including automated and manual recovery testing.
When did you last test your backups?
Whether you’re an existing Weald IT client or looking for advice on backup and disaster recovery, we’re happy to review your arrangements and discuss the options available, including automated and manual recovery testing.
Tags In
- Your Backups Are Working. But Can You Actually Recover?
- Lessons We’ve Learned: Recruiting Good People
- Your IT Is Working — But Is It Working Well?
- Could Your SharePoint Files Suddenly Become Read-Only?
- When an Employee Leaves, Is Their IT Access Really Removed?
- You Have MFA. But Is It the Right MFA?
- IT Support Pricing Changes from 1 January 2027
- Lessons We’ve Learned: Why We Don’t Chase Rapid Growth
- Lessons We’ve Learned Using AI in a Real Business
- Microsoft 365 Price Changes from 1 July 2026