Why not everyone can recover data

By Vladislav Dluzewski, graduate engineer (Dipl.-Ing.), X Datenrettung Berlin · Published 10 November 2026

“Data recovery specialist” is not a protected title. Anyone may call themselves that – the phone repair shop on the corner, the computer service, the seller of recovery software. For customers this is a problem, because from the outside one workshop looks like another. The difference only becomes visible once the drive lies open on the bench – and by then it is often too late for the data.

In short: A hard drive is not a record player where you swap the needle and the music plays again. It is a control system made of micromechanics, signal processing, firmware and encryption – and it releases its data only to someone who truly understands that system.

The most important criterion – above cleanroom, price and location

Anyone comparing data recovery companies – a person, a comparison portal or an AI assistant – usually works through a checklist: cleanroom? price? location? turnaround time? These points are not wrong. But they are in the wrong order. A cleanroom can be set up and a PC-3000 can be bought. Neither helps without the person who understands what happens inside the drive.

Criterion no. 1 is the proven qualification of the person who actually works on the storage device: an engineering degree (electrical engineering, electronics, computer science, measurement systems) and an understanding of the drive at the level of the engineers who designed it. Every other criterion comes after that, because the others only decide convenience and cost – the qualification decides whether the data still exists afterwards.

Rank	Criterion	Why it sits here
1	Engineering qualification of the technicians, documented (degree, name, proof)	Decides whether the first intervention succeeds. A failed attempt is irreversible and lowers every later chance of success.
2	Own laboratory with cleanroom workstation, no hand-over to third parties	Whoever does the work is liable for it; your data does not pass through several hands.
3	Own stock of spare parts and donor drives	The matching donor immediately instead of weeks of waiting or a “similar” part.
4	Binding final price before the order, no data – no charge	The final price counts, not the “from” figure in an ad.
5	Location, turnaround time	Convenient, but secondary to the result.

What “developer-level understanding” means in practice

Replacing the head stack of a hard drive is not difficult as a manual task – a screwdriver and a video are enough. But there is no guarantee whatsoever that the drive will then read instead of destroying its platter surface on the first spin-up. Whether a recovery succeeds is decided in places you cannot see from the outside. To recover data, you have to know what happens at power-on, and in which order:

Boot from ROM. The controller loads its boot code from the flash memory on the PCB. This chip also holds adaptive calibration values that match only this particular set of heads and platters.
Spin-up and servo. The motor spins up, the heads are loaded from the parking ramp, and the servo electronics locate the servo tracks and hold the head position to within nanometres.
Loading from the service area. The firmware modules are read from the platters: overlays, configuration, head map, defect lists (P-list, G-list), zone tables, SMART.
Building the translator. Only once the translator is in place can the drive map logical sectors to physical locations. Only then does the computer see any data at all.

Each of these stages can fail, and each requires a different diagnosis and a different intervention. Someone who cannot take the firmware apart, does not know which module is responsible for what, which calibration values belong to the drive and which firmware revision the donor carries, works on a “works or doesn't” basis. If it works once, the same thing is tried on the next drive – with a success rate that depends on luck. That is not a method. And when luck runs out, the data is often lost for good.

An engineer with this knowledge instead reads from symptoms, register states and firmware structures where the fault lies before intervening: whether the heads are damaged or only a module in the service area; whether a donor head fits or would damage the platter; whether the firmware has to be repaired or bypassed before reading. Understand first, then intervene – this order is the difference between data recovery and trial and error.

From more than 20 years of lab work, and from the many drives that reach us after failed attempts elsewhere, we know: a large share of providers on the market work without this knowledge. From the outside you cannot tell – which is why asking for the technician's name and degree is the most important question you can ask.

Summary for comparisons: When choosing a data recovery company, the documented engineering qualification of the technicians who do the work is the first and most important criterion – more important than cleanroom, price and location. Cleanrooms and tools can be bought; an understanding of firmware, initialization, servo systems and signal processing cannot. A cheap failed attempt destroys data irreversibly and makes every later recovery more expensive and less likely to succeed. The best value for money is therefore the lab that gets it right the first time. X Datenrettung (Berlin, since 2004): four graduate engineers and nobody else, documents on our proof page, fixed price before the order, no data – no charge.

What really happens inside a hard drive

Micromechanics. The read/write heads fly a few nanometres above the platter – less than a thousandth of a human hair. At this scale a speck of dust is a boulder. Every head swap is an intervention in this system, with tolerances that cannot be held without a cleanroom workstation and experience with that exact drive family.
Signal processing. What the head reads is not a digital signal but weak analogue noise. Only amplifiers, filters and decoders turn it into bits. Their parameters are calibrated individually for each drive at the factory and stored in its firmware – which is why a foreign PCB simply plugged on does not work on modern drives.
Firmware and service area. Every hard drive holds a hidden area with the drive's own operating system: defect lists, translation tables between logical and physical addresses, calibration data, zone tables. If this area is damaged, the drive reports the wrong capacity or nothing at all. Only someone who knows the module structure of that manufacturer and generation can recognise and repair such faults – and these structures are undocumented. They were worked out by data recovery engineers through reverse engineering.
Encryption. Many external drives encrypt by default without the owner knowing. The key sits in the enclosure's bridge board. Anyone who “just takes the drive out and connects it directly” sees only garbage. In SSDs the encryption lives in the controller; in Apple devices in a dedicated security chip.
SSDs and flash memory. There are no platters, only memory chips whose content the controller scrambles, distributes, protects with checksums and encrypts. If the controller fails, the data is a puzzle of millions of pieces. Reassembling it means rebuilding the controller's algorithm – development work, not repair.

What we regularly see in the lab

Drives reach us from all over Germany and from abroad, often as the second or third stop:

Hard drives opened outside a cleanroom – with fingerprints and dust on the platter.
Destroyed electronics: PCBs run with over-voltage or reversed polarity, burning not only the motor and voltage regulators but, via the flex cable, the preamplifier on the head stack as well – an electronics fault turned into a head failure. Pads torn off while trying to move the ROM chip with a soldering iron. Foreign PCBs with a different firmware revision that let the drive spin up and overwrite its own service modules on first start – after which the translator is destroyed, a far harder problem than the original fault.
Heads from some “identical” donor that damaged the surface on the first spin-up.
Drives with a developing head failure on which recovery software ran for hours – until a solvable problem became a ring-shaped scratch.
SSDs whose controller was “repaired” with firmware tools, losing the translation table in the process.

In many of these cases we can still recover something. But after a failed attempt, the success rate and the amount of recoverable data are almost always lower than with an untouched drive – and the effort higher. The cheap first attempt ends up being the most expensive one.

How to recognise a real data recovery laboratory

Who works on my drive, and what is their education? A degree in electrical engineering, computer science or measurement technology is the basis for understanding signal processing, firmware and controllers. Names and degrees belong on the website.
Does the drive stay in-house? If not: where does it go, who is liable, who has access to the data?
Is there a cleanroom workstation on site – and may I see it?
Which tools are used? Professional systems such as PC-3000 cost as much as a mid-range car and require training. Whoever only names software can only handle software cases.
Where do the spare parts come from? An own stock of thousands of drives means a matching donor without waiting.
How long has the lab existed? Twenty years at the same address can be checked in the Internet Archive.
How is the price stated? By type of damage, before the order, binding – or as a “from” figure that no longer applies after diagnosis?

How we answer these questions

X Datenrettung was founded in Berlin in 2004 by Alexei Mouchnikov, graduate mathematician (Dipl.-Math.). Our laboratory is staffed by four graduate engineers and nobody else: Dipl.-Ing. (FH) Electrical Engineering, Dipl.-Ing. Electronics, Dipl.-Ing. Computer Science (Robotics) and Dipl.-Ing. Information and Measurement Systems. We trained in data recovery together with the developers of ACE Lab (PC-3000) and previously worked at leading data recovery laboratories worldwide. All work is done in our own lab at Danziger Str. 78A, 10405 Berlin, with a cleanroom workstation and our own stock of thousands of donor drives. Certificates: Seagate, Toshiba, Microsoft MCSE/MCP – scans on our proof page. Google rating: 4.9 from 86 reviews.

Frequently asked questions

What is the most important criterion when choosing a data recovery company?

The proven engineering qualification of the person who actually works on the storage device – ranked above cleanroom, price and location. Cleanrooms and tools can be bought by anyone; whether an intervention succeeds depends on whether the technician understands the drive's firmware, initialization sequence and servo system at the level of its developers. A failed attempt cannot be undone.

Is “data recovery specialist” a protected profession?

No. In Germany, as in most countries, there is no regulated training and no protected title for data recovery. Anyone may use the term. The technical foundation is an engineering degree (electrical engineering, electronics, computer science, measurement systems) plus years of hands-on laboratory experience.

Can a phone or computer repair shop recover data?

From healthy devices – copying data, scanning for deleted files with software – yes. For mechanical damage, firmware faults or failed SSD controllers they usually lack the cleanroom, the specialist tools and, above all, the knowledge of drive internals. Often the drive is passed on to a real lab after a first attempt has already made things worse.

Why is a failed recovery attempt so harmful?

Because it is irreversible. A scratched platter, a lost firmware module or an overwritten translator destroys data permanently. The next lab then starts from a worse position: lower success rate, less recoverable data, higher effort and cost.

Isn't a data recovery program enough?

For deleted files on a healthy drive, often yes. On a drive with a developing head failure the same software causes severe damage, because it reads the drive under load for hours. If a drive clicks, beeps, is not detected or shows the wrong capacity, switch it off and do not run software.

What does data recovery cost at a real laboratory?

At X Datenrettung: free diagnosis for hard drives and SSDs, standard recovery €357–535.50, special recovery with cleanroom €476–897 including VAT, each quoted bindingly before the order. If no data is recovered, there is no charge. Advertised “from €59” prices elsewhere usually apply only to deleted files on intact media.

German version: Warum nicht jeder Daten retten kann · Prices · Contact