A MacBook rarely makes a fuss on the way out. It stops accepting the charger, or the lid gets closed and it never wakes again, or a folder with a question mark turns up halfway through a coffee in the Lexicon. Where the storage can be unplugged, which covers the SATA disks in the older bodies and Apple's own blade SSDs from 2013 through to 2017, the balance of probability is firmly that everything written to it is still there, spilled drinks included.
Anything sent in for MacBook 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.
On a MacBook the keyboard sits directly over the board, so a spill reaches the electronics in seconds and starts corroding them the same day. Drying it out and hoping is the expensive route. On the models where storage is a removable blade, get the blade out and posted while the corrosion is still young.
A dead board is not a dead disk, and on a pre-2018 machine those two are entirely separable components. Plenty of MacBooks arrive here written off by a repair shop with storage that images without a single retry. Establishing which part failed is free and takes 2 working days from booking in.
macOS has searched for a system to start and found nothing usable. On the older machines that means a failing disk or a torn file system, both of which recover. On anything with a T2 or an M-series chip it means a conversation about what is realistically possible, which is a shorter conversation.
A macOS update interrupted by a failing sector leaves the machine looping through a screen that gets no further. The user data is untouched. It is read off an image of the disk rather than by persuading the installer to have another go.
MacBook Pros up to 2012 hold an ordinary 2.5-inch SATA disk, and those fail with the same click as any other. Donor heads in filtered air, then a slow clone. The chassis stays with you and only the drive needs to travel, which keeps the postage sensible.
The proprietary blades in Retina Pros and Airs stop without any warning at all. One boot they are there, the next the machine cannot see them. They go to the flash bench for service-mode work, and where the controller will not respond at all, for a direct read of the NAND with the mapping rebuilt afterwards.
Disk Utility offers erase next to first aid, and on a small trackpad in a hurry those are easy to confuse. An erase writes a new structure over the old one rather than scrubbing the contents. What determines the result is how much has been written since, so shut it down now.
No obstacle at all when the password or the recovery key comes with the machine. The disk is imaged while it is still sealed and the credentials are applied to the copy, which keeps every unlock attempt away from hardware that may only have a few hours of reliable running left in it.
macOS keeps local Time Machine snapshots, and they are frequently the fastest route back to a document deleted this morning. They are also the first thing macOS discards when the disk fills up. If something has just been deleted, stop saving anything else to that machine immediately.
The screen is black and the machine is running perfectly well underneath it. That is a display fault, not a storage fault. On a machine with a removable blade the answer is to take the storage out and read it here. On a soldered machine it is a display repair, which is a different trade to ours.
Every SSD keeps a table translating the addresses your Mac uses into the physical NAND behind them. Corrupt that table and a perfectly healthy set of chips becomes unreadable in an instant. Rebuilding it is specialist flash work rather than anything a piece of software can do from a USB caddy.
Ribbon cables in these chassis chafe where they fold, and a worn SATA or blade cable produces exactly the symptoms of a dying disk: disappearing volumes, stalls, a machine that will not boot. It is worth ruling out first because it costs nothing and it is more common than it should be.
A swollen pack lifts the trackpad and bows the base, and on some models it presses directly against the storage. Stop charging it, stop carrying it, and get the data off before the swelling does anything more interesting than making the machine wobble on a desk.
Liquid finds the trackpad gap and travels along the ribbon into the board. This is the quietest spill there is, because nothing sounds wrong for a fortnight, and then contacts start corroding through. Boards from these jobs go through an ultrasonic bath before anything is imaged.
Second opinions are ordinary work here. The storage is imaged as it now stands, whatever was attempted previously, and the assessment describes what is genuinely readable. It is fair to say that a machine that has already been apart twice is a harder job than one that has not.
Aluminium heats quickly and the machine has nowhere to shed it. Nothing fails on the day. The damage shows up later as a blade that has started dropping off the bus when the machine gets warm, which is a symptom people spend months blaming on software.
Repeated panics can be storage, memory or a graphics fault, and diagnosing them properly takes longer than simply securing the data first. The order that has never once produced a worse result is this one: image the disk, then argue about the cause.
Prepared for sale, erased, then the sale fell through and it turned out the photographs had not been copied anywhere after all. On the older machines with plain disks there is usually a great deal left. On anything with hardware encryption there is not, because erasing the keys is instantaneous and final.
On a MacBook of 2012 or earlier this matters a great deal, because a spinning disk that hits the floor gets head damage. Later machines have no moving parts in the storage at all, so a drop tends to break the display, the case or the board instead, and the flash is often fine.
Something left on the keyboard, the lid shut on top of it, and the pressure passes straight into the board underneath. The display cracks first and the logic board takes the rest. Storage is rarely the casualty, which is why the machine can be a write-off while the files are perfectly safe.
Third-party USB-C supplies that skip the negotiation stage put voltages onto a board that it was not designed to see. Board repair takes 50% of the quote upfront and sits outside no fix, no fee, and on a soldered machine a dead board is the end of the road for the data.
A standard NVMe stick and an adapter is a cheap way to enlarge an older MacBook, and it introduces its own faults: sleep that never wakes, volumes that vanish after an update, firmware that never anticipated this particular controller. The module is read here whatever adapter it arrived wearing.
Holding the power button down to end a session that has stalled is a write interrupted mid-flight, every time. Do it often enough and the file system accumulates small inconsistencies until one of them lands somewhere structural. Rebuilding is done from an image, not from the live volume.
A well-documented board fault that kills the machine outright while leaving the hard disk in perfect condition. These arrive regularly, usually after being told the data was lost with the machine. It was not. The disk unscrews, travels on its own and images normally.
An EFI or firmware password locks the machine at boot and it cannot be cleared by resetting anything. On the older models this is an inconvenience rather than a barrier, because the storage comes out and is read directly. On T2 and Apple Silicon machines it is a barrier, and an absolute one.
Storage on these is soldered to the board and encrypted by the T2, with the keys held inside the chip itself. The pads on the board are a service interface for Apple, not a recovery route, and reading the flash directly returns nothing but ciphertext. These machines are declined, plainly and in advance.
The flash sits in the same package as the processor and is bound to the Secure Enclave. While the board runs, the data is reachable through the machine. Once the board is dead, the data is gone, and no laboratory anywhere can change that. We would rather say so now than take a fee to confirm it later.
The Windows side vanishes after a macOS update far more often than Apple documentation suggests, taking the NTFS volume with it as far as the Mac is concerned. The partition table is usually intact underneath. Both file systems are read from one image and the files come back from whichever side they were saved on.
Vibration, damp and temperature swings between a cold morning and a warm cab age everything inside a laptop faster than a desk ever would. On the older MacBooks the disk fails first. On the newer ones it is usually the board, and on a soldered machine that is the whole story.
Reflowing or hot-air work on a board with the storage still fitted is a good way to turn one fault into two. Where the blade has been heated, the NAND may still read but the controller often will not. That is flash work at the chip level, which carries 50% upfront and falls outside no fix, no fee.
The deciding factor is the storage inside, not the year printed on the receipt. Anything up to 2012 carries a 2.5-inch SATA disk, and that is as ordinary as Apple work gets. Enquiries about MacBook Pro hard drive repair almost always turn out to concern one of those. Between 2013 and 2017 Apple used its own proprietary blade SSD, which is equally straightforward provided the correct adapter is already on the shelf, and that generation lies behind nearly every MacBook Air hard drive repair call this office takes. Both types release from a connector, get imaged, and are returned. Then 2018 arrives with the T2 chip and Apple Silicon extends the same design further. On those machines the flash is soldered into the logic board and protected by hardware encryption, which means a dead board keeps the contents, and no firm being straight with you will claim otherwise. Attempts on that hardware came off the service list roughly two years ago because too few of them succeeded to justify billing for the effort. To identify which era you have, the model number is printed in small characters on the underside of the case.
A spill reaches the logic board before it reaches anything else, and the board takes the damage that would otherwise have found its way to the storage. Take the disk or blade out of a machine in that state, clean and treat it, and it will usually image without complaint. Where these go wrong, optimism rather than chemistry is usually to blame. The charger gets connected just to see whether it survived. It spends a fortnight in a bag of rice. It sits on a radiator overnight to dry. Off the power, unplugged, and moved on quickly is the whole of the method, and getting those three things right settles most of a MacBook file recovery job before the parcel has even been labelled.
MacBook storage has changed shape roughly every three years, and the bench has to keep up with all of it:
Every pinout Apple has shipped, from the early Air blades through the Retina Pro years, plus plain SATA for the 2012 machines and earlier. The right adapter is on the shelf rather than ordered in.
Firmware access for spinning disks and service-mode work for blade controllers. Where a controller will not respond, the NAND is read directly and the translation layer is rebuilt from what comes back.
Spill boards are neutralised, cleaned ultrasonically and dried until they present steadily. Nothing is imaged through an intermittent connection, because a capture that keeps dropping does more damage than no capture at all.
Containers, snapshots, volume groups and catalogue records read straight from an image, with earlier consistent states used where the current one is damaged.
Erased containers and reinstalled systems are searched at block level for surviving records. Slow work, and on a mechanical disk it is frequently the difference between nothing and everything.
Machines that have already been somewhere else are imaged as they stand. The report says what is readable now rather than what might have been readable before somebody else opened it.
The dividing line is 2018. Before it, the disk or the blade unclips, travels on its own in a padded envelope, and is recovered as routine work at £300 + VAT a unit, with a free diagnostic that closes 2 working days after booking in and a finished recovery running 2 to 4 working days. From 2018 the T2 fuses encrypted flash to the logic board, and Apple Silicon puts it inside the processor package, so once the board has died nothing readable remains for anybody to work with. Spill survivors, frayed cables imitating dead drives, blades that have mislaid their own mapping, boards that died while the storage stayed in perfect health: every one of those ends up as a module under a bench light and then as files on new media. Bracknell to Guildford is about forty minutes on the A322 and the A3 if you would rather hand it in yourself.
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.