Data Recovery Case File · Formatted & Logical Faults · The Percentage Question

"What percentage will you recover?" — the honest trade answers after imaging, in a manifest with names on it: bad sectors and delays decoded, the repair tool's defeat welcomed, and 930GB reconciled line by line

His enquiry ended on the question every owner wants to ask and few do: "can you advise the likelihood of successfully recovering the files, to what extent — i.e., percentage of data recovered — and how much this would cost?" The case around it: a Seagate 1TB external holding "essential personal files, totalling ~930GB"; every attempt to enter it met "the 'corrupted and unreadable' error." His efforts were the reasonable ladder: the system's disk-repair tool, run "unsuccessfully"; the platform's own file-recovery utility, likewise; and a third-party scan whose findings he reported precisely — "Bad Sectors and Delays on the drive." This page answers his percentage question with the only version that isn't salesmanship — and decodes why his two failed tools failed, why one of those failures was quietly the best news in the enquiry, and what bad-sectors-plus-delays was really measuring.

The decode: what his scan measured, why the tools lost, and the honest shape of a percentage

Bad sectors and delays, read together: the third-party scan handed him the diagnosis without the translation. Bad sectors are patches of the recording surface that no longer read; delays are their outriders — sectors that still answer, but only after the drive's internal retries. Together they describe media decay in progress: a surface developing weak territory, with the filesystem's critical structures unlucky enough to sit on some of it. That is why the volume greeted him as "corrupted and unreadable" — the paperwork wasn't scrambled so much as resting on ground that no longer holds it up.

The two defeats, ranked: the file-recovery utility failed for the ordinary reason — it must read through the same failing surface as everything else, and stalls where the surface does. The disk-repair tool's failure deserves warmer welcome: that tool writes — it repairs paperwork by amending it — and repair-writes against decaying media are how salvageable volumes get amputated, structures "fixed" into forms that orphan the very files they indexed. Its refusal to complete was the system declining to operate on a patient it couldn't hold steady. In this archive's ledger, that failed repair is the best thing that didn't happen to his 930GB.

The percentage question, answered honestly: here is the industry truth his question deserves. A confident percentage quoted before assessment is marketing — no one has measured anything yet, and precise-sounding numbers are chosen for comfort, not accuracy. The honest sequence inverts it: image first, with an error-tolerant pass that maps every sector as read, retried, or lost — then the percentage stops being an estimate and becomes a manifest: this file verified whole, this one recovered with a flaw at such-and-such offset, this one resting on dead ground. Likelihood, in this fault class, could be stated fairly in advance — decayed-media cases with intact mechanisms recover well, typically overwhelmingly — but the extent arrives as a document with filenames on it, or it isn't an answer at all.

The recovery: mapped, rebuilt, and reconciled by name

The drive was imaged natively with error-tolerant handling — healthy territory harvested at speed, weak sectors retried with patience, the losses mapped to exact addresses. On the image, the filesystem's structures were reconstructed clear of the decayed ground, and the volume stood up with its folders and names. Then the answer to his question, in its honest form: a per-file manifest across the ~930GB — the overwhelming majority verified whole; a small population recovered with mapped flaws, each identified; and a short list, named, whose territory had genuinely died. Essential files: present and verified, every one.

On the bench

The manifest this page promised is machinery, not rhetoric. Data Extractor imaged the decaying surface error-tolerantly while building its sector map in real time — every address logged as read, retried or lost — and the filesystem was reconstructed on the image clear of the weak ground. OSForensics then walked the recovered estate file by file: each opened or hash-verified, each cross-referenced against the sector ledger, producing the per-file account — verified whole, recovered with a mapped flaw, or resting on dead sectors — that turned his percentage question into a document with his own filenames on it. Firms quote percentages; this is what a bench delivers instead.

The outcome

The estate delivered on new media with the manifest beside it — his percentage question answered to one decimal place after the measuring, which is the only time the decimal means anything. Free assessment, one fixed written figure including VAT, no recovery, no fee. And the portable lesson, filed for every reader who asks his excellent question: firms quote percentages; benches deliver manifests. Ask which one you're being offered.

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.