One day, all of this will be gone. The sun will swell up and consume the inner planets and burn it all to a crisp. Luckily, all life on Earth and everything humanity has created on this planet will be dust for millions of years before that happens, so that’s a bullet dodged!
I hate to depress you on such a nice day, but entropy is inevitable, and that counts for your backup strategies too. All backup strategies fail eventually, so the goal isn’t to aim for the impossible, but to plan for reality.
The myth of the “perfect” backup
You might feel that you’re following the perfect backup strategy, such as the hallowed “3-2-1” approach, where you have multiple copies on multiple media in multiple p…
One day, all of this will be gone. The sun will swell up and consume the inner planets and burn it all to a crisp. Luckily, all life on Earth and everything humanity has created on this planet will be dust for millions of years before that happens, so that’s a bullet dodged!
I hate to depress you on such a nice day, but entropy is inevitable, and that counts for your backup strategies too. All backup strategies fail eventually, so the goal isn’t to aim for the impossible, but to plan for reality.
The myth of the “perfect” backup
You might feel that you’re following the perfect backup strategy, such as the hallowed “3-2-1” approach, where you have multiple copies on multiple media in multiple physical locations. However, no plan survives contact with the enemy, and the enemy here is Murphy’s Law. Which is to say that you’re sleeping well at night because you think your backup strategy is bulletproof—until it’s actually put to the test.
Credit: Andrew Heinzman / How-To Geek
All you’re really doing is reducing the probability that you won’t lose your data, but drives die, cloud systems fail, NAS arrays can become corrupted, and sometimes your “redundancies” end up being the same mistake, but repeated a few times.
No matter what, entropy or human complacency will get that data in the end.
Hardware doesn’t care about your data
Every form of data storage has an expiration date. So far, it seems the best way to store data over long periods of time are clay tablets or carving them into rock, and even then you need some luck with the weather and erosion.
“Hot” storage—running drives—fail from usage. A mechanical drive might have an MTBF (Mean Time Before Failure) of over a million hours, but it could fail a few days out of the box, or never during its useful lifespan.
SSDs wear out from being written to, optical discs pressed at a factory can suffer from disc rot, the dyes in burned discs are doing well if they survive ten years. Magnetic tape is one of the best long-term formats, but it can suffer from tape degradation, or you simply may have no way to read it down the line.
SSDs and hard drives used for “cold” storage can fail without being turned on. The hard drive’s lubricant and rubber components may degrade and fail, and flash memory can suffer from “bit rot” as the electric charge leaks from memory cells over many years without being powered on.
Age will kill all media in the long term, but a power surge or a virus can take out your date right now. The only way that redundancies help is if they’re truly independent of all your other backups, which for most home backups isn’t the case.
The cloud isn’t a failsafe—it’s someone else’s computer
Credit: Lucas Gouveia/How-To Geek | Chantal de Bruijne/Shutterstock
Cloud storage may have given us a false sense of security. Yes, cloud storage providers follow industry-standard data storage practices to make it unlikely your data will be lost. However, we’ve all experienced the various major cloud service outages over the past few years, like the 2025 Amazon Web Services outage that took down a significant percentage of the web’s most popular services.
When an outage like that happens, you have to reckon both with the realization of how much of your stuff you suddenly can’t access and with the reality that those servers may not come back online again some day. Often the cause is simple human error, or malicious actors, but it can just be bad luck. Apart from this, in the long run the service could go under financially, or you might miss a few payments. Either way, it’s not really your backup of the data, is it?
The human factor breaks everything
Credit: Fractal Pictures/Shutterstock.com
We weak fleshy humans are a common reason backup strategies fail, whether in the cloud or at home. Did you test and verify your media and files? Did you set up a schedule and then check if it was done? What about simply mislabeling stuff? That’s how you lose a cool billion dollars in cryptocurrency (via NPR). There’s no automation that can make up for our mistakes or lack of due diligence.
The only winning move is constant paranoia
The best backup strategy is one with a healthy dose of paranoia, and a realistic goal for how long that data must be protected. Do you need it safe for five years? Ten? The rest of your life? The rest of your children’s lives? These questions lead to very different backup practices and choices.
While it’s a stressful way to live, if you have truly irreplaceable and important information, you need to have proportional levels of effort. True redundancy, periodic migration to fresh media, and frequent verification that your data is still intact.
Every backup strategy will fail—your job is to make sure it doesn’t fail all at once.
UGREEN NASync DH4300 4-Bay NAS
CPU Rockchip RK3588C (2.4GHz)
Memory 8GB LPDDR4X
Drive Bays 4
OS UGOS