Data recovery occupies an unusual place in computer repair. It's an area surrounded by unrealistic expectations in both directions — people who assume that any lost data can be recovered with the right tools, and people who assume that once a drive stops working, everything on it is gone forever. The reality is more nuanced, and understanding it can help you make better decisions when something goes wrong.
This article explains what data recovery actually involves, the different types of failure and how they affect recoverability, what factors influence the outcome, and — critically — what you should and should not do if you find yourself in this situation.
The fundamentals: how data is stored
Understanding the basics helps make sense of why some recoveries succeed and others don't.
Traditional hard disk drives (HDDs) store data on spinning magnetic platters. Read/write heads float a tiny distance above the platter surface and read or write data by detecting or changing the magnetic polarity of small regions. The heads never touch the platters during normal operation; if they do — known as a head crash — the result is physical damage to the platter surface.
Solid state drives (SSDs) store data in flash memory cells. There are no moving parts. Data is retained by trapping electrons in a floating gate structure. SSDs are generally more resistant to shock and mechanical failure than HDDs, but they have their own failure modes — particularly around controller failure and the finite write endurance of the memory cells.
When data is "deleted" from most file systems, the actual data on the storage medium is usually not immediately overwritten. Instead, the file system marks the space as available for reuse. The data remains until new data is written to those sectors. This is why recently deleted files can often be recovered — as long as the drive hasn't written significant amounts of new data since the deletion.
Types of failure and what they mean for recovery
Logical failure
Logical failures are problems at the file system or software level, rather than the physical hardware. The drive itself is intact and functioning; the problem is that the data can't be accessed or located correctly.
Examples include: corrupted file system (the drive's index has become damaged and the OS can no longer read it); accidentally deleted or formatted partitions; accidental deletion of files; ransomware encryption (the files exist but their contents have been scrambled); and OS corruption that prevents booting.
Logical failures generally have the highest recovery success rates, because the underlying data is often still intact on the drive. Specialist recovery software can often reconstruct the file system or directly scan for recognisable file signatures, bypassing the damaged index.
Electronic / firmware failure
The drive's controller board or firmware can fail independently of the storage medium. In these cases, the platters or memory chips containing the data may be completely undamaged, but the drive can't communicate properly with the computer or access the data correctly.
Recovery in these cases typically involves replacing the controller board (matching the exact revision and sometimes transferring firmware chips) or using specialist tools to bypass the compromised firmware. Success rates are reasonably good when the underlying medium is undamaged.
Mechanical failure (HDD)
Mechanical failures in HDDs are the most serious category from a recovery standpoint. This includes failed read/write heads, a seized spindle motor, and damaged platters. The characteristic clicking sound of a drive with failed heads is a well-known warning sign.
Recovery in these cases typically requires a cleanroom environment — a controlled space where drives can be opened without contaminating the platters with dust particles. Even a small particle on a platter surface can cause additional head crashes and worsen the data loss. This type of work is beyond what most local repair shops can handle; it requires specialist equipment and facilities.
We are transparent about this: severe mechanical failures with HDD head damage are outside the scope of what we can address directly, and we'll tell you this clearly. In these cases, a specialist cleanroom data recovery service is the appropriate route, though the cost reflects the complexity of the work.
Physical damage
Physical damage to the storage medium — whether from impact, fire, flood, or head crashes — is the most challenging scenario. The prognosis depends entirely on the nature and extent of the damage. Flood-damaged drives that have not been powered on since the event sometimes yield good recovery results when handled promptly; drives that have been operated after physical damage often make things significantly worse.
SSD-specific failure modes
SSDs fail differently from HDDs. Common failure modes include: controller failure (the chips that manage the drive stop functioning, even though the flash memory is intact); firmware corruption; worn memory cells (after many write cycles, cells lose their ability to hold charge reliably); and sudden power loss during a write operation, which can corrupt data structures.
One important characteristic of SSD failure: they often fail more abruptly than HDDs, with less warning. An HDD may produce click sounds or slow read times before failing; an SSD may simply stop responding with little prior indication. This makes backups particularly important for SSD-based systems.
Factors that affect recovery success
How much the drive has been used since the problem
This is probably the most significant controllable factor. Every write operation to a failing drive potentially overwrites data that could otherwise be recovered. If a drive has become inaccessible or shows signs of failure, the most important thing to do is stop using it. Don't attempt to run chkdsk, don't defragment it, don't install recovery software to the same drive.
For accidental file deletion on an otherwise healthy drive, the same principle applies: stop using the drive immediately and seek recovery help before any further data is written.
The nature and severity of the fault
As outlined above, logical and electronic failures generally have better recovery prospects than severe mechanical or physical damage. A straightforward deleted files recovery has a very different expected outcome to a drive that's sustained head damage.
Whether previous recovery attempts have been made
Well-intentioned but ill-advised recovery attempts can significantly reduce the likelihood of a successful professional recovery. Using consumer "undelete" tools on a failing HDD, running the drive for extended periods when it's clearly in distress, or attempting to open the drive outside a cleanroom environment can all cause additional damage.
Drive age and pre-existing condition
A drive with a long history of SMART errors is in a more precarious state than one that was in good health before the failure. Pre-existing bad sectors, for example, mean that some data may have been corrupted or relocated before the current event.
What to do when a drive stops working
The most useful actions are also the simplest:
Stop using the drive. Power off the computer immediately. Don't attempt to run diagnostics, reinstall the OS, or download recovery software to the same drive.
Don't repeatedly power cycle a failing HDD. Each time a drive with failing heads spins up, there's a risk of further platter damage. Power it on once for assessment, if at all.
Keep the drive safe. Store it at room temperature in a dry, stable environment. Don't place it near magnets, in direct sunlight, or anywhere it might get bumped.
Get a professional assessment before deciding anything. The assessment will give you a realistic picture of what's recoverable and at what cost — and that information should drive the decision about whether to proceed.
Recovery success depends heavily on acting quickly and correctly in the first few hours. The most common thing that reduces the chances of a good outcome is continued use of a failing drive.
Realistic expectations
We'll always tell you honestly what we think is achievable before proceeding. Recovery rates vary enormously depending on the scenario — from very high (simple logical failure on an otherwise healthy drive) to very low or zero (severe physical damage, heavily overwritten deleted data).
We don't charge for a recovery that yields no usable data. The assessment is free; if we determine that recovery isn't feasible, or if the recovered data doesn't contain what you needed, you won't be invoiced for recovery work.
And beyond recovery: the experience of losing data — or nearly losing it — is usually the most effective prompt to take backups seriously. If you're not running a reliable backup routine and would like advice on setting one up, get in touch and we'll point you towards a straightforward solution.