Data Recovery Case File · Second Fixes & Trade Handoffs · One of the Genuine Walls
A Verdict of NAND Degradation: Verifying a Real Limit
The enquiry was straightforward and did not overstate its case. "We've got an SSD which has failed and we'd like the data recovering. We've taken it to another company and they couldn't get the data back due to severe NAND degradation. Not sure if you'd be able to recover the data at all, or if it's not recoverable." Most second-opinion enquiries in this archive arrive after a verdict that turns out to describe the limits of somebody's equipment. This one is different, and honesty requires saying so upfront: NAND degradation is one of the genuine walls in this trade, not a euphemism. The previous firm may simply be right. But the phrase is also used loosely, and there is a specific distinction worth checking before accepting it.
| Media | Failed solid-state drive assessed elsewhere — recovery unsuccessful with severe memory degradation cited as the cause |
| Reported situation | Drive failed in service · prior recovery attempt unsuccessful · degradation of the memory reported as the limiting factor · owner uncertain whether any route remains |
| Fault class | Reported memory-level degradation — distinguishing genuine cell-level failure from unsuccessful controller access determines whether any route exists |
| Equipment used | Prior verdict verified rather than accepted · vendor technological and safe-mode access attempted (PC-3000 portfolio) · error rates measured directly where memory was read · position and limits reported in writing |
The decode: what degradation is, and the distinction worth checking
What NAND degradation actually means: flash memory stores data as trapped electrical charge, and cells wear with use — every write cycle degrades the insulation slightly. Worn cells hold charge less reliably, so the values read back drift, and the drive's error correction compensates until it cannot. Beyond that point the data returned from those cells is genuinely unrecoverable: it is not hidden or locked, it has ceased to be reliably stored. That is a real physical limit, and unlike most verdicts in this archive it does not yield to better equipment. Where a drive has genuinely reached it, no firm can help, and one that promises otherwise is not being straight.
Why the verdict still deserves checking: because the phrase is used to cover two different findings. The first is what it says: the memory was read, the error rates were measured, and correction was exhausted. The second is broader and more common — controller-level access failed, the drive could not be brought up, and degradation was inferred rather than measured. Those sound identical to a customer and are entirely different technically, because a drive whose controller cannot be accessed may still have perfectly healthy memory behind it.
How the two are told apart: by what was actually done. If the memory itself was read and its error characteristics measured, the verdict is a measurement and should be respected. If the attempt ended at the controller — no vendor-mode access achieved, no firmware area assessment, no read of the memory — then degradation was a conclusion rather than a finding. The useful question to put to the previous firm is simply: was the memory read, and were the error rates measured? A straight answer either way is worth having.
What a second look consists of: attempting vendor technological and safe-mode access, assessing the firmware area, and — where access is restored — reading the memory and measuring its condition directly. That either produces data, or it produces a measured confirmation of the original verdict, which has value of its own: knowing the answer is genuinely no is better than wondering.
The honest framing: this is offered as verification rather than as a promise of a different outcome. The assessment costs nothing, and if the first verdict was right, that is what will be reported.
On the bench
The prior verdict was treated as a hypothesis to verify rather than a conclusion to accept — and equally, not as a verdict to overturn, since memory-level degradation is one of the genuine limits. Access was attempted through the PC-3000 portfolio's SSD support in the drive's vendor technological modes, with the firmware area assessed for the corruption that commonly prevents start-up. Where access was restored, the memory was read and its error characteristics measured directly rather than inferred, which is the only way to distinguish genuine cell-level failure from an unsuccessful attempt at the controller. The position and its limits were reported in writing either way.
The outcome
The verdict verified by measurement rather than accepted or contradicted, and the position reported in writing. 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 given this verdict: NAND degradation is a real physical limit, because worn flash cells hold charge unreliably and once error correction is exhausted the data has genuinely ceased to be stored — so unlike most verdicts, this one may simply be correct; but the phrase also gets used where controller access failed and degradation was inferred rather than measured, and a drive whose controller could not be brought up may have healthy memory behind it. Ask the previous firm one question: was the memory read and were the error rates measured?
Told your SSD has degraded memory
Take it seriously — this is one of the genuine limits rather than a fob-off. Flash stores data as trapped charge in cells that wear with use, and once those cells hold charge unreliably enough to exhaust the drive's error correction, the data has stopped being reliably stored. No equipment recovers that. But ask one specific question before accepting it: was the memory actually read and its error rates measured, or did the attempt end at the controller? Those produce identical-sounding verdicts and are completely different, because a drive whose controller couldn't be brought up may have perfectly healthy memory behind it, and degradation would then be an inference rather than a finding. A firm that measured it will say so plainly. If you seek a second opinion, ask for it framed as verification rather than as a promise — and be prepared for a measured confirmation of the first answer, which is still worth having.
Worth verifying, not overturning — call Glasgow Data Recovery on 0141 404 0294; vendor-mode access attempted, error rates measured directly where the memory reads, and the position reported in writing.
Request a quote online →
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.