Data Recovery Case File · Trust, Practice & Honest Limits · The List Before the Decision

Can I See What Is Recoverable First?

His enquiry asked for something entirely sensible that customers rarely think to request. An mSATA solid-state drive that has stopped working: "I think there may be some physical defect preventing correct operation. I would be interested in receiving a list of directories and files found, to see what can be recovered. Would you be able to provide a cost for recovery and a timescale?" Wanting to know what is actually there before committing money is a good instinct, and on a great many jobs it can be honoured. But the honest answer has a shape, because whether a file list can be produced before you commit depends on what is wrong — and on some faults the list is the output of the expensive part rather than something that precedes it.

The decode: three situations, three different answers

Where a list can be produced first: if the drive still responds — it identifies itself and can be read, even slowly or with errors — then it can be imaged, and the filesystem read from the image, and a directory listing produced. That listing costs nothing beyond the assessment and is genuinely useful: he can see what exists, decide whether it is worth the figure, and identify priorities. On logical faults, damaged filesystems and struggling-but-readable drives, this should be normal practice and is worth asking for.

Where it cannot: if the drive does not respond at all, there is nothing to list. A device that does not identify itself yields no filesystem, no directory and no inventory, because the work that would make those visible is the recovery. On a failed controller or a drive that will not become ready, a firm offering a file list up front is describing something it cannot do until after the expensive part. That is not evasion; it is the order the work happens in.

Where his drive probably sits: solid-state failures cluster at the unresponsive end. When an SSD's controller fails, the drive typically presents nothing at all — no capacity, no model, nothing to enumerate — because the controller is the entire management layer rather than merely a doorway. So the realistic sequence for him is: free assessment to establish whether the drive responds at all; if it does, an image and a listing before he commits; if it does not, vendor technological-mode access first, with the listing following once the drive can be addressed.

What to ask for instead, when a list is not possible: two things that are. A written statement of the fault class and the realistic outcome, so the decision rests on something. And a staged commitment — access attempted first, with the listing produced and reviewed before the remainder of the work is authorised. That gives him most of what he wanted without asking anyone to invent an inventory.

On his other two questions: the cost should be one fixed figure including VAT issued in writing before any chargeable work, and the timescale should accompany it rather than be estimated verbally. On solid-state work the honest caveat belongs alongside: controller-level access frequently succeeds and sometimes genuinely does not, and reading the memory packages directly is not the dependable fallback it is on cards and sticks, because the address mapping and often the encryption live inside the controller.

On the bench

The fault class was established free of charge before any sequence was proposed, since it determines whether a listing can precede the work or must follow it. Where the drive responded, it was imaged and the filesystem read from the image, and a full directory listing was issued to him before any chargeable work — so that the decision rested on what actually existed. Where it did not respond, access was attempted through the PC-3000 portfolio's SSD support in vendor technological modes, with the honest limits set out in writing first and the listing produced as soon as the drive could be addressed.

The outcome

The fault class established at no cost, the listing produced where the drive allowed it, and the sequence stated in writing before commitment. Free assessment, one fixed written figure including VAT; where a chip has to be removed, 50% of parts and labour is payable upfront with the balance only on success — otherwise no recovery, no fee. The decode, for anyone who wants to see the files first: it is a fair request and often achievable — a drive that still responds can be imaged and listed before you commit, and on logical faults you should ask for exactly that; but where a drive does not identify itself there is nothing to list, because the work that makes a filesystem visible is the recovery, and a firm promising an inventory up front on a dead drive is describing something impossible. Ask instead for a written fault class and a staged commitment.

Our case files are drawn from genuine enquiries received by our laboratory over the past ten years, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery procedure our engineers apply to that fault, using the equipment listed.