Request a quote
Emergency SectorsCareers About us Blog Get in touch
Technician working on an opened laptop, with a screwdriver on the desk

Data destruction: wipe or destroy the drive?

Data destruction sounds like one job, and it is really a decision you make per drive. Shredding feels safest, but it throws away resale value, a small SSD can survive a coarse shredder, and without a certificate you cannot prove either. Here is how to decide drive by drive, and what to record so it stands up to an audit.

7 min read Written by the ITproposal team
Short answer

Deleting and formatting leave the data in place. Wiping is a method that covers the whole drive and is then verified.

Wiping keeps the drive usable and is usually enough for a working disk; destruction is for failed drives and for data your policy says must be destroyed.

The proof is a certificate per drive, with the drive’s own serial number, the method, the verification, the date and who did it.

Why deleting and formatting are not data destruction

Deleting a file removes the pointer to it; the data stays on the disk until something happens to overwrite it. A quick format does much the same for the whole drive. Free recovery tools get a lot of it back, and that is exactly what happens when a laptop finds a new owner through an online marketplace or a forgotten storeroom.

There are two ways to make data genuinely unrecoverable: sanitise the drive with a method that reaches every location where data can sit, or destroy it so it can no longer be read. The widely used NIST categories put it neatly: clear (overwrite), purge (erase in a way that defeats laboratory recovery) and destroy. In the UK, the NCSC publishes guidance on secure sanitisation of storage media along the same lines.

Under UK GDPR, your organisation stays responsible for personal data after the device has left the building. If a drive turns up later with customer records on it, that is your breach to report to the ICO, not the collector’s.

Wipe or destroy: three questions that decide it

The choice is less a matter of principle than it looks. Three questions settle it for almost every drive:

  • Does the drive still work? A drive that no longer responds cannot be wiped or verified. It always goes to destruction.
  • Is the device going back into use? A laptop a few years old often has value or a second life, but only with its drive. Destroy the drive and you destroy part of that value. For reuse, verified wiping is the method.
  • What does your own policy say? Some organisations require physical destruction for certain data, such as health records or protectively marked information, whatever the condition of the drive. Then that is the decision, not a technical trade-off.

If a drive falls into none of the strict cases, verified wiping is usually enough. Write that rule into your policy so that every disposal does not start the same discussion again.

And check retention before you wipe. A finance laptop or an old server may hold records you still have to keep for tax or contractual reasons. Archive first, then wipe; the other way round does not work.

An SSD is not a hard drive, and a printer is a storage device too

Overwriting works well on a traditional hard drive: every sector is written and then checked. On an SSD it works badly, because the controller decides where data lands and keeps spare capacity you cannot address. There, the manufacturer’s built-in sanitise command is the right route, or crypto-erase if the drive was encrypted from day one. Degaussing does nothing to an SSD.

The storage most often forgotten at disposal:

  • Printers and multifunction devices with an internal drive or memory full of scanned documents.
  • NAS units and servers with several drives in a RAID set: each drive is its own item with its own serial number.
  • Phones and tablets, where a factory reset is only safe if the device was encrypted and is then removed from device management.
  • USB sticks, memory cards and backup tapes, which rarely appear in any asset register.
  • Firewalls, switches and access points, holding configurations, passwords and sometimes certificates.

Destroying a drive: what “broken” has to mean

A drive is only destroyed when nobody can read the data back. Drilling a hole or taking a hammer to it makes the drive unusable but leaves pieces of the platters intact, and afterwards you cannot show what was and was not hit.

Professional destruction uses a shredder that reduces the drive to particles of a fixed maximum size, chosen to match the sensitivity of the data. The German standard DIN 66399 is a common reference for this, with separate classes for hard drives and for electronic media. That distinction matters: an SSD or memory chip is small enough to pass through a coarse shredder intact. So ask not only whether drives are shredded but to what particle size, per type of media.

On-site destruction lets you watch it happen. Off-site destruction moves the risk to transport, which is what the next section is about.

The proof: what a certificate of destruction must show

A certificate of data destruction is the only thing you still hold once the drives are gone. An auditor, the ICO or your own data protection officer will look at that document, not at good intentions. What it should contain:

  • The serial number of the drive, not just the device. A laptop may have had its drive swapped, and then the certificate is about the wrong disk.
  • The method and the standard: overwrite, built-in sanitise, crypto-erase or physical destruction, and to which standard.
  • Verification: whether the result was checked, and how. “Wiped” without verification is a claim.
  • Date, place and who did it, under whose responsibility.
  • The chain before it: when the drive was collected, by whom, and whether the counts at collection and at processing match.

Match the certificates against your asset register. The device that is in the register and on no certificate is exactly the one to chase. How we apply each method per type of media is on our page about data destruction and secure wiping.

The weak point comes before the wipe

Most drives do not go missing during wiping but in the weeks before it: in a box under a desk, in a cupboard anyone can open, in the boot of someone who is “just dropping it off”. Three arrangements help:

  • One place where retired equipment is collected, locked, with a list updated every time something goes in.
  • A fixed interval between handing in and wiping, so a year’s stock never builds up. The longer equipment sits, the less it is worth and the likelier something is missing.
  • Counting at handover: whoever collects counts by serial number and signs. Discrepancies are resolved there and then, not in the final report.

Wiping on site removes the transport risk; wiping or destroying off site is often more practical for large volumes. Both are defensible as long as the chain of custody is complete on paper.

Where our work stops

What we do. We collect equipment, record every drive by serial number, wipe using a recognised method or arrange physical destruction where wiping is not possible or not allowed, and provide a certificate per drive. We hold ISO/IEC 27001 certification, and that is how we run the chain.

What we do not do. We do not decide how your data is classified or how long you must keep it; that follows from your own policy, your retention obligations and the law. We are not lawyers or a regulator, and a certificate from us is not a judgement on whether your organisation complies with UK GDPR. Protectively marked government information has its own rules and its own approval route, and a supplier does not set those.

Frequently asked

Questions we get about this

The five that come up most often.

Is formatting a hard drive enough to destroy the data?

No. A quick format only removes the pointers, and free tools can recover much of the data. What works is sanitising the whole drive with a recognised method and verifying the result, or physically destroying it.

Can an SSD be wiped securely or does it need to be destroyed?

A working SSD can be wiped securely with the manufacturer’s built-in sanitise command, or by crypto-erase if it was encrypted. Simple overwriting works poorly on SSDs and degaussing does not work at all. A failed SSD goes to destruction, using a shredder fine enough for memory chips.

What is the safest way to destroy a hard drive?

Shredding to particles of a fixed maximum size, matched to the sensitivity of the data. Drilling or hammering makes the drive unusable but leaves fragments intact and cannot be evidenced afterwards. Make sure the destruction is recorded per serial number.

What should a certificate of data destruction include?

The serial number of the drive itself, the method and standard used, whether and how the result was verified, the date and who carried it out. It should also cover the chain of custody: when the drive was collected and whether the counts at collection and processing match.

Can you tell us which method we are required to use?

No. The required method follows from your own policy, how your data is classified and your legal obligations, and that is your organisation’s decision. We carry out the chosen method, record it per drive and explain the technical differences between the options.

Drives still waiting to go?

Tell us what media you have and whether any of it still has resale value. You will get a proposal with the method per type of drive and the certificate that goes with it.

Practical IT knowledge in your inbox

New guides on management, security and the workplace, written by the people doing the work. No sales talk, and you can unsubscribe in one click.

We use your address for the newsletter only. See our privacy policy.

Request a quote Call