What makes these jobs painful is the timing. The machine gives up on the morning a deadline lands, or starts rebooting in circles on the one evening set aside for copying the photographs somewhere safer, or meets a mug of tea before nine o'clock. Quite possibly the laptop is beyond saving. On the evidence of this bench, whatever it was storing generally is not, and the storage is where your work actually resides. Machines reach us from Bracknell, Wokingham, Camberley, Sandhurst and the whole string of towns running down the A322.
Anything sent in for laptop recovery is diagnosed free. The written figure that follows is settled before a screwdriver leaves the drawer: £300 + VAT for any single hard drive or SSD, NVMe included.
No fix, no fee covers logical recoveries. Outside it sit electronic and mechanical failures, chip-level work, DVR jobs and forensic jobs, and physical work is 50% up front. Every band is published on the data recovery cost page.
Reading a symptom back to the failure underneath it is where the work genuinely starts, and this set of thirty accounts for all but a handful of the parcels that reach the Guildford bench. If yours is not on the list, it will still be recognised — describe it on the phone and you will get a straight answer about the odds before you post anything.
Charging boards, DC jacks and power circuits fail on their own timetable and have nothing whatever to do with where your files are kept. A good share of the laptops written off in Bracknell households this year were carrying storage in perfect health. The only question worth answering first is which of the two components actually stopped, and that is settled on a bench in under an hour rather than argued about over the phone.
Heads that were flying a few nanometres above a spinning platter have just been introduced to it. What you hear afterwards is a rasp or a repeating tick, and every further attempt to boot drags damaged heads across more of the surface. This means matched donor parts fitted in filtered air. Mechanical work sits outside no fix, no fee and takes 50% of the quote upfront.
Better odds by a wide margin. With the drive at rest the heads are parked off the platter on a ramp, so a knock in that state is frequently survived outright. Treat it as a reason to take one careful image rather than a reason to relax, because shock damage often surfaces a fortnight later as a patch of unreadable sectors in the folder you use daily.
Drying the chassis on a radiator settles nothing. Liquid leaves salts and acids behind and those carry on eating pads and tracks for weeks after the machine looks dry. Storage lifted out within a day or two nearly always images cleanly. Storage left in the case until the following month usually needs the board neutralised and bathed before it will present steadily.
The panel is the expensive part and the storage is the important one, and the two are unrelated. Take the base off, remove the disk or the M.2 stick, send that on its own. There is no sense paying to post a broken shell down the A322 when the part that matters weighs sixty grams.
A puffed cell pushes against whatever sits above it, and on plenty of chassis that is the drive bay or the M.2 slot. Two steps in this order: stop charging it today, and get the storage out this week. A swollen pack sitting in a house is a fire risk as well as a recovery problem, which is worth rather more urgency than it usually gets.
Windows cannot find a working copy of itself. That happens when a disk is failing and it happens when a driver update goes badly, and separating the two matters less than you might expect, because the route out is identical. The drive is imaged in full and the documents come off the image, never off a system that has already fallen over.
A stalled feature update wrecks the installation and leaves everything belonging to the user untouched underneath it. Provided the disk itself reads properly, this is among the most straightforward jobs described anywhere on this page and rarely the one that worries us.
That message is the drive's own firmware admitting it has run out of spare sectors to hide damage in. It is a genuine warning and it is usually the last one you get. Shut the machine down where it stands, take the disk out, and let nothing further write to it.
Seagate's 1TB shingled laptop drive is the most common mechanical failure crossing this bench. Read speed collapses first, the machine hangs for minutes on end, and a soft tick arrives later. Imaging hardware handles it by measuring which heads still perform, taking the strong regions first and coming back to the weak ones with a map already built.
A click every few seconds is the actuator sweeping out, finding nothing it recognises and resetting itself. That is a head or preamp fault and no amount of restarting improves it. Donors are matched on model and firmware revision from catalogued stock, fitted in filtered air, and the clone is taken at whatever pace the drive is willing to tolerate.
Dell, HP and Lenovo fit whichever module was cheapest that quarter, so the stick behind the panel is usually an SK hynix BC711, a Samsung PM9A1 or a WD SN530 rather than anything the owner chose. When the controller stops answering there is no noise and no warning. The module goes to the flash bench for a service-mode session, and where that fails, for a direct read of the NAND with the translation layer rebuilt afterwards.
The honest answer, which most of this trade avoids giving. TRIM tells the drive to clear deleted blocks in the background, and once it has, nothing on any bench anywhere brings them back. Where the deletion was recent and the machine was powered down quickly there is sometimes a little left. Where it has been in daily use since, expect very little and be pleasantly surprised rather than let down.
Most people searching for this have a perfectly healthy drive and a mislaid key, which is not a laboratory job at all. Check the Microsoft account the laptop signs into, under devices. On a work machine the IT team holds it in Entra or Active Directory. It may also be on a printed sheet from setup day, in a saved text file, or on a USB stick in a drawer. If the key turns up, you are finished and you owe us nothing.
Nearly always the right drive and the wrong key. Every BitLocker protector carries an identifier and the volume names the one it is waiting for, so a key belonging to another machine, or to an earlier encryption of the same machine, is refused without any explanation offered. Send the disk with every key you have been able to find and the matching gets done here.
The machine is quick again and eleven years of photographs are sitting underneath the new installation. A clean install writes to a small fraction of a disk and leaves the rest legible. What decides how much comes back is how heavily the laptop has been used since. The single most useful thing you can do is stop using it.
On a spinning disk, deleting removes the reference and leaves the contents in place until something else claims the space. On an SSD, read the note about TRIM further up this page. Either way the correct next move is the same one: power it down and put nothing new on it.
Some machines pair a small fast module with a large slow disk and hand Windows a single volume through Intel RST. Break the small half and the large one presents as unformatted, which is alarming and quite inaccurate. Both parts are imaged, the pairing is unpicked in software, and the volume reads properly again from the copies.
Storage soldered to the mainboard is outside what this lab takes on, and that is worth saying before you pay for postage rather than after. On those machines the flash is part of the board. There is no module to unscrew, no adapter that helps, and no honest promise available to make about them.
A blocked fan does not kill a laptop in an afternoon. It runs every component twenty degrees warmer for years, and in most chassis the drive bay sits directly downstream of the exhaust. By the time read errors start appearing, the fan has been struggling for so long that nobody thinks to blame it.
Glass, sunlight and no airflow will take the inside of a laptop well past anything the manufacturer ever tested for. Nothing appears to happen at the time, which is the trouble with it. The bill arrives some weeks later in the form of sectors that will not read.
Unbranded USB-C bricks and barrel-jack supplies take out charging circuits, and now and then they take the drive electronics with them. Both are routine work. Board-level repair is invasive, so it needs 50% of the quote upfront and falls outside the no fix, no fee guarantee.
An intermittent DC jack cuts writes off midway at random intervals, so the file system degrades in small instalments rather than in one memorable event. The machine grows steadily odder over months and then stops. Everything is rebuilt from a complete image and nothing is repaired on the live disk.
Consumer cloning software has exactly one strategy for a bad sector, which is to ask for it again, and then again. Left running overnight against a failing drive it will wear out heads that still had useful work left in them. What arrives afterwards is in worse condition than it was, and there is usually still more readable than the owner has been told to hope for.
Two things wrong with the folklore. Rice does nothing at all to liquid that has already reached the board, and warm air drives the residue deeper into connectors as it dries. The useful action is mechanical rather than ritual: get the storage out, get it into a padded envelope, and do it on day one instead of day ten.
What Windows calls a shutdown is normally a partial hibernation with the lights off. Take that disk out and mount it on another computer and the file system is describing a moment that has already passed, which is how a rescue attempt becomes structural damage. Work here happens against an image, with the dirty state handled properly rather than ignored.
That retaining screw goes astray during upgrades far more often than anyone admits to. The stick then sits at a slight angle, contact comes and goes, and writes get cut off mid-sentence. It presents exactly like a dying SSD and it is a fifty-pence part. The free diagnostic tells the two apart and closes 2 working days after the module is booked in.
A temporary profile means Windows could not load the account and has improvised a blank one instead. Nothing has been removed. The real profile folder is still on the disk exactly where it was, and it gets copied off directly rather than through an operating system that has stopped co-operating.
Half the files made the journey, the old drive was quick-formatted ready for resale, and the space between those two events is where the phone call comes from. A quick format rewrites the index and leaves the contents in place, so there is a great deal to read underneath it. The less that disk is touched from now on, the more of it comes back.
Second opinions are ordinary work rather than an awkward favour. The drive is imaged as it now stands, whatever was done to it previously, and the assessment describes what is genuinely readable rather than what ought to have been. Send any paperwork along with it, because it occasionally explains a pattern of damage that would otherwise take half a day to work out.
Laptop data retrieval is a storage job rather than a laptop job, and Toshiba makes the point better than any other badge on the sheet. Satellites from the C, L and Pro lines are old enough now that their original 2.5-inch mechanisms are simply reaching the end of a normal working life, which is why Toshiba Satellite recovery is one of the steadier lines here rather than an occasional curiosity. The Tecra and Portege business machines fail the same way and arrive from offices instead of spare rooms. What gives out is nearly always an MQ01ABD or MQ04ABF100 mechanism, and it gives out by degrees: reads that used to be instant start taking seconds, the machine hangs on the Toshiba logo, and a tick appears once a head has finally stopped reading. Everything after the first tick makes matters worse, so the sensible response to that noise is the power lead out and the base panel off. Removing that panel is your side of the job and not ours, because complete machines are never taken apart at this address; one door and four screws is usually all that stands between you and a bare drive in a jiffy bag. Later Toshiba laptops are sold under the Dynabook name and behave no differently. Whichever it is, the drive alone travels and the shell stays with you.
Asus splits cleanly into machines this bench can help and machines it cannot, and knowing which one you own before you buy a stamp is worth more than anything else on this page. A ROG gaming laptop or a mid-range VivoBook holds an M.2 module that unclips in about ten seconds, and Asus laptop data recovery on those is ordinary flash work: the module comes out, the controller is approached through the manufacturer's service mode, and where it refuses to answer at all the memory is read directly and the translation rebuilt in software. Budget VivoBook modules are frequently DRAM-less designs, which are cheap, slow and inclined to stop responding without any warning whatsoever. The thinner ZenBooks are the problem. A good number of them carry their flash soldered to the mainboard, and soldered flash is outside what gets recovered here by anybody, so an honest answer on the phone saves you the postage. That single question decides more Asus data recovery enquiries than any fault does, and it is settled in a minute: ring 0800 689 0668 with the model number off the underside before a screw is turned.
Lenovo arrives in three shapes and each one lands on a different bench. Older ThinkPads, the T and X series in particular, still carry a 2.5-inch mechanical disk, so Lenovo hard drive recovery on those is plain disk work: donor heads fitted in filtered air where a head has died, a slow image taken afterwards, and the awkward regions left until last. Current ThinkPads, IdeaPads and Yogas run M.2 NVMe modules instead, which makes Lenovo SSD recovery a controller and firmware problem, and a silent one, since a module that has lost its translator reports nothing at all rather than failing loudly. Some IdeaPads use the short 2242 modules, which are easy to miss under a heatsink and just as easy to unclip once found. The exception is the thinnest Yoga and IdeaPad Slim models, where the storage is part of the board and cannot be lifted off cleanly. Dell, HP, Acer, MSI and the Surface range all take the same route through this bench, and the figure does not move with the badge: £300 + VAT for the drive, after a diagnostic that costs nothing.
An impact that leaves the shell looking perfectly fine can still be enough to drop the heads onto the platters, which is why a machine that ticks after an accident, or that stalls at the manufacturer's logo and progresses no further, should be switched off and left that way. The real damage comes with the second and third attempt rather than the first. Liquid works to a different timetable altogether. Corrosion begins the moment anything gets inside, and no household remedy has ever won that race. Rice absorbs a negligible amount. A radiator evaporates the water and leaves the residue behind, baked on. Pressing the power button to find out whether it survived is behind a substantial share of the genuine write-offs that arrive here. Have the storage out quickly, or get somebody else to do it, and the odds stay well on your side.
A laptop bay might hold a 2007 IDE disk or a Gen4 NVMe stick, so the bench covers both ends of that span:
Each unit is photographed and logged the day it lands, handled on an earthed mat, and copied end to end before recovery work starts. That order is not rearranged for anybody, because it is the order that stops a weak drive becoming a dead one.
SATA in 2.5-inch and mSATA, M.2 in both SATA and NVMe forms, the short 2242 and 2230 sticks, Apple blades sorted by year, ZIF and IDE going back two decades. Whatever the machine was built with plugs in natively rather than through something improvised.
Firmware-level access for spinning disks, service-mode sessions for SSD controllers, and direct NAND reads where a controller cannot be persuaded to talk at all. Chip-level work counts as physical work, so it carries 50% upfront and sits outside no fix, no fee.
Head stacks are exchanged inside a laminar-flow bench and never on an open desk. Every donor on the shelf is logged by model and by firmware revision, because a part that matches on capacity alone is no use at all and is not fitted.
Spill boards are neutralised, bathed and dried until the drive presents steadily on the bus. Imaging waits until that point, because a connection that comes and goes during a capture does more harm than no connection at all.
Read-only is enforced in hardware from the moment a customer drive is connected. No stage of this workflow is physically capable of writing to media somebody sent us.
Fleet Latitudes and EliteBooks out of the offices along the A329(M), Swifts and Aspires that have been carried between Bracknell and half the universities in the south, Legions from spare rooms in Binfield, and the Pavilion that has lived under a bed in Bullbrook since 2017. The badge on the lid matters far less than the part behind the panel. If it spins, expect an ST1000LM035 or a WD10SPZX. If it does not, expect an SN530, a BC711 or a PM9A1, and expect no warning noise of any kind. Send the bare drive or module and hold on to the laptop: a single drive or SSD is £300 + VAT, the diagnostic costs nothing and closes 2 working days after the unit is booked in, and a finished single-drive recovery runs 2 to 4 working days. Two things this lab will not do, said here rather than after your parcel has arrived. Flash soldered to a mainboard is declined, and phones and tablets are not accepted in any condition.
Acer sells hard to students and to households keeping an eye on the budget, so Swifts, Aspires and Nitros turn up here in numbers and turn up with hard lives behind them. Halls of residence, lecture halls, the back of a car, a kitchen floor. Inside an older Aspire is a plain 2.5-inch disk that fails audibly and needs donor heads in filtered air. Inside a current Swift is an M.2 module, often a budget DRAM-less design, and that gives no warning at all before it stops answering the bus. Those go to the flash bench for service-mode work or a direct read of the NAND. Both routes end the same way, with the storage out of the machine, an image taken, and the dissertation back on fresh media. The only Acers refused are the thinnest models, where the flash is part of the board.
Two ranges that arrive from opposite directions. Dell comes in from business, mostly Latitude 5000 and 7000 fleets retired in batches and mobile Precisions from design and survey firms, and the storage is almost always a standard M.2 module behind a base plate that lifts off in a minute. XPS machines are tighter inside but no different in principle. HP arrives from both sides, ProBooks and EliteBooks out of offices around the A329M and Pavilions out of spare rooms, and its 2.5-inch generation is old enough now that the mechanisms are simply reaching the end of their working lives. Neither maker locks storage to the machine, so a dead motherboard on either is a cheap outcome rather than an expensive one: the module or the disk comes out, the shell stays with you, and the figure is £300 + VAT after a free check. The exception on both is the thinnest chassis, where flash sits soldered to the board and no promise can honestly be made.
Fujitsu left the UK laptop market years ago and its machines keep arriving anyway, mostly Lifebooks retired out of office fleets and Amilos rescued from box rooms, with a decade of paperwork on them that exists in no second copy anywhere. The 2.5-inch mechanisms inside decline on a schedule you could nearly set a calendar by. Firmware hangs come first, then heads losing their edge, then sector damage arriving in patches. Take the drive out, send that alone, and the rest is ordinary mechanical work: the fragile regions are left until the end of a capture, and donors are fitted only once a head has actually died rather than as an opening gambit. Nothing else needs to travel, so the Lifebook can stay on whatever shelf it has been living on.
The storage has to come away from the machine before the parcel is made up, because a bare drive is the only thing that should reach Guildford. If the screws defeat you, ring 0800 689 0668 and it can be talked through on the phone; failing that, most repair counters charge next to nothing for ten minutes with a screwdriver.
The disk or the M.2 stick needs to be out of the machine before it travels, because laptops are not stripped down at this end. On most models that means one base panel and six or eight screws, and any repair shop will do it for a few pounds if you would rather not. Put the bare unit in an anti-static bag, add a slip carrying your name and telephone number, and send nothing else in the box. It goes tracked and insured to Guildford Data Recovery, Building 2, Ground Floor, Guildford Business Park, Guildford GU2 8XH, roughly forty minutes down the A322 and the A3 if you would sooner hand it in at reception, which is open 9:00am to 5:30pm, Mon–Fri. There is no collection service in Bracknell or anywhere else, so the journey is your side of the job.
Nearly every job here arrived as a parcel. Tracked, insured post is the calmest way to move a drive that is already struggling, and something handed over in Berkshire, Surrey or London is normally on the Guildford bench the next working day.
Is the storage still bolted into a machine — laptop, tower, iMac, MacBook, rack server, a DVR under the till? Free it first and send the bare unit. Stripping hardware is not something this lab does, though it is ten minutes' work for any repair shop on your high street. There is a single case with no way round it: memory chips soldered flat onto the mainboard, which is how Apple Silicon machines and certain ultrabooks are built. Where the storage cannot be unbolted, there is no parcel to make up.
↓ Print the shipping & booking-in form (PDF)
The name on the parcel wants to be Guildford Data Recovery. Driving it over from Bracknell is roughly forty minutes on the A322 then the A3; posting it costs you a stamp and a day. Either way, a message goes out to you as soon as it is logged onto the system, and two working days later the diagnostic is finished.
Unsure whether something should go in the box? Ring 0800 689 0668 while the lid is still open, or work through the free online diagnostic and let it tell you.
Nothing to pay for the diagnosis, one written figure before any work begins, and the band on this page is £300 + VAT for any single hard drive or SSD, NVMe included.