Most disaster recovery plans are sound right up until someone tests a restore. If the backup is missing, incomplete or itself corrupt, the data is still usually recoverable from the failed hardware — and that is the work we do.
We are a recovery lab, not a managed backup provider — so this page is about the other half of the plan: what happens when the restore doesn’t work, and how the data comes back off the hardware that failed.
The backups that fail tend to fail quietly. A job that has been reporting success for months turns out to have been excluding a mapped drive; a NAS replicating to a second NAS in the same cupboard is taken out by the same power event; a database backup runs while the database is open and captures an inconsistent snapshot. None of that shows up until a restore is attempted under pressure, which is the worst possible moment to discover it. The other frequent gap is scope: virtual machines and mail stores are backed up, and the shared drive everyone actually works from is not.
We recover data from hardware that has failed — servers, RAID arrays, NAS units, virtual machine stores, databases and individual drives. Every device is imaged read-only before any reconstruction, arrays are rebuilt virtually from those images rather than through the controller, and nothing is written back to the original media. Work starts from £500 +VAT for RAID, NAS and server systems after a free 48-hour diagnostic and a written quote, and most jobs are no fix, no fee. Where a case needs drive-level physical work, we take a 50% deposit and the balance is due only if we succeed.
What we don’t do is sell you a backup product. There are good providers for that, and we would rather be honest about where our part starts.
The first is separation: a copy in the same building, on the same power feed, or on the same network as the original is exposed to the same event, and ransomware in particular follows the network to whatever it can reach. The second is testing: restore something real, at random, on a schedule — a backup that has never been restored is an assumption. The third is the immutable or offline copy, because that is the one thing a ransomware run cannot encrypt on its way through.
We work with businesses across Oxfordshire and the surrounding counties — Oxford, Banbury, Bicester, Didcot and Abingdon, out to Swindon, Reading, Aylesbury and Northampton. Kit can be dropped off at John Eccles House on the Oxford Science Park or sent in by insured post — the diagnostic and terms are identical either way.
Power it down and leave it down. Don’t rebuild, don’t re-initialise, and don’t restore a backup over the top of the original volume until you know the data is safe elsewhere. A failed restore that writes to the source is one of the commonest ways recoverable data is lost after the initial failure.
Usually, yes. A failed backup doesn’t change the state of the original hardware, and data on a failed server, RAID or NAS is generally still recoverable provided nothing has been written over it since. The free 48-hour diagnostic tells you what is realistically retrievable before you commit to anything.
No — we are a data recovery lab. We recover data from failed hardware and can advise honestly on where a backup arrangement is exposed, but we don’t sell backup software, managed services or cloud storage. That keeps the advice straight.
Server, RAID and NAS recovery starts from £500 plus VAT, confirmed in writing after a free 48-hour diagnostic. Single drives are from £300 plus VAT. Most jobs are no fix, no fee; drive-level physical work carries a 50% deposit with the balance due only on success.
The diagnostic is free and normally back within 48 hours, and straightforward array work typically completes in a few working days from approval. Where donor parts or extensive imaging are needed it takes longer, and we will say so at the quote rather than after.