Archiving isn’t the same job as working storage, and the drive that’s best for your editing scratch disk is often the wrong choice for a box of files you’ll touch twice a decade. The core question is simple: what happens to your data if the drive sits powered off in a drawer for three years? SSDs and hard drives fail differently, and that difference matters more for archives than almost any other storage decision you’ll make.
As an Amazon Associate we earn from qualifying purchases at no extra cost to you.
How each one actually fails
Hard drives are mechanical. The failure modes are physical: bearing wear, head crashes, stiction on startup after long dormancy, and eventually just old age on the motor. A drive that’s been spinning in a NAS for five years straight tends to fail gradually, often with warning signs in SMART data (reallocated sectors, rising seek errors) before it dies outright.
SSDs fail differently, and the failure mode that matters for archiving specifically is charge leakage. NAND flash stores data as electrical charge in cells, and that charge slowly drains over time. Powered off and sitting on a shelf, a consumer SSD is only rated to retain data for about one to two years at room temperature, and that spec gets worse, not better, as the drive gets older or as the NAND wears down from use. A brand new drive stored in a cool, dry place will likely hold data far longer than the JEDEC minimum spec, but “likely” isn’t a word you want attached to your only copy of ten years of family photos.
HDDs, by contrast, don’t rely on stored charge. A hard drive that’s been sitting unplugged for five years will almost certainly spin up and read fine, assuming it was healthy when you put it away. The main long-dormancy risk is stiction, bearings seizing after years of non-use, which is a real but less common failure than people assume, and it’s one you can often work around (gentle tapping, trying it in a different enclosure) in a way you simply can’t work around decaying NAND charge.
Side by side for archive duty
| Factor | HDD | SSD |
|---|---|---|
| Cost per TB | Lower, roughly $15-20/TB for desktop drives | Higher, often 3-5x the per-TB cost |
| Unpowered data retention | Years, limited mainly by mechanical seizure risk | 1-2 years typical, shorter if drive is older or worn |
| Failure warning | Often gradual, SMART-detectable | Can be gradual or sudden, less predictable |
| Physical shock sensitivity | High when spinning, low when off | Very low, no moving parts |
| Best use case here | Primary archive medium | Secondary copy, active archive you access often |
Note the last row is doing real work here: this isn’t “SSDs are bad,” it’s that each one covers a different failure mode. An SSD won’t die from being bumped off a shelf. A powered-off HDD won’t lose data to charge leakage. A good archive strategy uses both.
What to actually buy
For the bulk archive, a conventional desktop hard drive is the right call for most people, and it’s not a compromise, it’s just the correct tool. Look at NAS-rated internal hard drives if the archive lives in a NAS or runs 24/7, since those drives are built for continuous vibration and extended duty cycles that desktop-class drives aren’t rated for. If the archive is going to sit in a drawer rather than spin constantly, a standard desktop drive in an external enclosure works fine and costs less.
Where SSDs earn their place is as the second copy, or as the drive you actually reach for regularly. If you’re archiving project files you still open a few times a year, a portable external SSD is faster, more durable in transit, and silent, and the charge-retention concern matters less because regular use (plugging it in, letting it refresh the cells) keeps the data healthy. The retention clock really only becomes a problem for drives that are powered on rarely or never.
If you’re buying bare drives to put in your own enclosure or NAS bay, don’t assume all drives of the same capacity are equivalent. Check whether you’re looking at 5400 RPM archive drives versus 7200 RPM drives, since the slower, cooler-running drives are often explicitly marketed for archival and cold storage, with lower duty cycle ratings but also lower cost and heat.
The rule that matters more than the drive type
Here’s the thing that gets lost in SSD-vs-HDD debates: no single drive, of either type, is a backup. The 3-2-1 approach (three copies, two different media types, one offsite) exists specifically because every storage medium fails eventually, just on different timelines and for different reasons. A $20/TB hard drive that you’ve actually verified by plugging it in once a year is more reliable than a single pristine SSD you never touch.
If your archive is genuinely irreplaceable, don’t spend your budget trying to find the “safest” single drive. Spend it on a second drive instead. Two hard drives from different purchase batches, one kept offsite at a relative’s house or in a safe deposit box, will outperform any one drive’s spec sheet by a wide margin.
Practical recommendation
For most people archiving family photos, documents, and old project files: buy two desktop or NAS hard drives, store one locally and one offsite, and power them up and spot-check the data once a year. Use an SSD for the files you access often enough that speed matters, and don’t let one sit unpowered for more than a year or two without plugging it in to refresh. The drive type matters less than whether you actually have a second copy somewhere else.
