They cost a few pounds, they live loose in a bag, and they hold the only copy of something irreplaceable. Here's why flash storage fails so suddenly — and what can still be saved.
There's a strange relationship between price and trust when it comes to storage. A £400 laptop is treated carefully and backed up. A £6 USB stick is carried loose in a coat pocket, plugged into borrowed computers, dropped in car footwells — and used to hold the only copy of a wedding video, a dissertation, or a business's entire accounts.
We see USB sticks, SD cards and microSD cards every week, and the pattern is consistent: the smaller and cheaper the device, the more irreplaceable the contents, and the fewer copies exist anywhere else. Flash storage has become the disposable drive that nobody treats as disposable.
The important difference
Mechanical drives usually give warning signs before they die. Flash storage often doesn't. It works perfectly one moment and is completely unrecognised the next — which is why a second copy matters so much more here than anywhere else.
A USB stick or memory card is not simply a small hard drive. It has no moving parts at all. It consists of memory chips holding the data, and a small controller chip that manages how they're read and written — and it's the controller, not the memory, that usually fails first.
Don't save anything new to it, don't reformat it, don't run a "repair" tool. Flash storage reuses space constantly, and every write reduces what can be recovered.
The familiar "you need to format this disk before you can use it" message is almost never the solution. Formatting rebuilds the file system and makes recovery considerably harder.
If it appears in one laptop but not another, note which. But try it once — not ten times. Repeated attempts to make the device appear can push a marginal controller over the edge.
We regularly receive sticks with cracked casings and gouged chips where somebody had a go. The memory chips inside are tiny and easily destroyed, and a broken package can put data out of reach permanently.
Flash recovery is often carried out without touching the original device at all, so there is very little to lose by having it looked at.
Where the problem is logical — a corrupted file system, deleted files, a drive that appears empty — the work is done entirely in software, and the original device is never modified. We image it first, then work from the image.
Where the controller has failed, normal software cannot read the device at all, because the software addresses the card through the controller that no longer works. In these cases the route is chip-off recovery: the memory chips are physically removed from the board and read directly, bypassing the controller entirely. The data has to then be reassembled — flash controllers spread and scramble data across chips in ways that vary between manufacturers, so this is a genuine technical exercise rather than a simple read.
It is not always possible. Monolithic drives, where the controller and memory are fused into one sealed package, are the hardest cases of all. But chip-off recovery succeeds far more often than people expect, and it is often the only remaining option.
A word about camera cards
If a card fails mid-shoot, stop using it straight away — and whatever you do, don't shoot over it. Photographers often reformat in the field to keep working, which overwrites exactly the data that was still recoverable. Put the card aside, use a different one, and deal with it afterwards.
If a stick or card has stopped being recognised, bring it in rather than experimenting with it. In flash recovery, the difference between a good outcome and a permanent loss is usually how much has been written to the device since it failed.
Don't format it and don't keep plugging it in. Bring it to us and we'll tell you honestly what's recoverable — with no fee if we can't get your data back.