Data Recovery Case File · Business & Research · The Partial-Recovery Question

One partition of two, priced honestly: why cost follows the fault, not the gigabytes — and a research lab's data brought back

The enquiry was institutional in the best sense: precise, budget-conscious, and already competently triaged. A laboratory computer failed to start after its power was switched off and on; the machine's built-in diagnostics declared the hard drive broken; and the in-house IT person made a proper attempt before stopping — reporting that he could see the drive only "on-and-off, making it hard to copy the data off it." The patient: a Toshiba 500GB from 2014, two partitions, carrying "a lot of lab data" the institution wanted back. And the closing question, asked the way procurement asks things: could we quote for recovering one partition of the drive? The question deserves the honest answer about how partial-recovery pricing really works — because the intuition behind it (half the drive, half the cost) meets an economics the trade should explain rather than exploit.

Two decodes: the on-and-off drive, and the one-partition economics

The intermittency first, because it vindicates the IT person's stop. A drive that appears and vanishes "on-and-off" is operating at the edge of its ability — completing its introduction on good attempts, dropping mid-session when its degraded firmware or heads stumble — and desk-side copying from that state is a lottery drawn per file: transfers die partway, each reconnection spends another marginal start-up, and (as another case in this archive documents at length) the copies that do complete from an unstable drive aren't automatically trustworthy. Recognising the lottery and declining to keep playing it was the professional move. Now the pricing question, answered plainly. What a recovery costs is driven overwhelmingly by the fault — the engineering required to make an unstable drive readable at all: diagnosis, firmware repair, any donor parts, and the managed imaging that must, for safety and coherence, effectively cover the drive as a whole, since a patient this marginal gets one careful supervised pass, not two visits. Against that core, restricting the target to one partition trims the genuine but modest margins: extraction scope, verification effort, delivery media. So the honest quote structure — and the one this enquiry received — shows both: the fault-driven figure that any recovery of this drive requires, and the true (small) saving that partition-selective delivery offers against full extraction. Half the gigabytes is not half the surgery; a lab that prices as though it were is either overcharging the full jobs or underdelivering the partial ones.

The recovery — whole-drive discipline, partition-selective delivery

On the PC-3000 the intermittency's authors were addressed directly — the age-worn firmware records repaired, the drive's dropout behaviour replaced by stabilised, managed access — and the full 500GB imaged in one patient pass, both partitions banked, because the image is also the institution's insurance against tomorrow's follow-up request. Delivery then honoured the procurement question exactly: the data partition — the laboratory's working files, years of research material — extracted from the image, verified by opening files across it with the confidentiality handling institutional data warrants, and returned on new media; the system partition held on the image through an agreed retention window (the "actually, could we also have…" that follows most partial recoveries costs nothing to anticipate), then securely disposed of on written confirmation. The quote had shown the honest arithmetic; the institution chose with it; the work matched it to the line.

Outcome

The research data recovered in full, the budget question answered without theatre — and the procurement guidance this case files for every organisation asking its version of "just the one partition." Ask for the decomposed quote: the fault-driven core, the delivery-scope margin, stated separately — it's the structure that lets a budget holder decide honestly, and the labs comfortable providing it are the ones whose full-job pricing survives daylight too. Credit the in-house triage in the enquiry, as this one did; an IT person's honest "intermittent, couldn't copy, stopping here" is worth a day of assessment. And bank the operational lesson beneath the whole case: a 2014 workstation drive carrying "a lot of lab data" in one place had been running on borrowed redundancy for years — the recovered set went back onto institutional storage with backup, which is the line item this entire page was always really about.

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.