If you’ve ever experienced data loss, you know the frustration it can bring. To keep such disasters at bay, it’s important to cultivate responsible practices for storing digital archives and backups. You could have the most well-organized collection of data in existence, but without backups in place, you’re only one accident away from disaster. Last week’s Tidy Tuesday post discussed building good habits for working with old and new data, and this week, we’ll extend that discussion to integrate safeguards into the way you work with data as soon as you receive or create it. This way, backups become proactive and automatic rather than an afterthought–or forgotten entirely.

There’s a decent bit of technical detail in this post, so before you get started, keep the following important point in mind:

A single backup copy is MUCH better than no backup, even if you don’t follow the conventional rules outlined below.

Some of you may already know everything I describe in this post. For others, it might seem overwhelmingly complicated and arcane. For anyone unsure how to proceed, I recommend that you obtain an external USB hard disk and then use either File History (for Windows users) or Time Machine (for Mac users) to start automatically backing up your most important files. This requires minimal effort while providing a significant improvement by leveraging built-in system software and a single inexpensive drive.

Archives vs. Backups: What’s the Difference?

Before we dive in, let’s clarify that “archives” and “backups” are not actually the same thing, even though you might think of the terms as similar or even interchangeable. An archive is comparable to a library in that it stores information and materials for long-term retention and occasional use. Conversely, backups create copies of data or documents for safeguarding against destruction or loss; these copies can be used to recover lost records in the event of data deterioration, accidental deletion, or malicious damage.

Here are two simple points to highlight the difference:

• Archives are where you move some of your data when you decide you no longer need to have it readily available for day-to-day use.
• Backups are duplicate copies of any data (archived or otherwise) that can prevent permanent loss in case of disaster.

You probably want both for a well-rounded organizational system. But even if you don’t feel a need to split your data between “active” and “archived,” you definitely want some kind of backup system in place.

It’s also worth mentioning briefly that syncing (data synchronization) services such as Dropbox are also not exactly the same as backups. While there is some overlap in what they do, and syncing services are certainly useful in many cases, they focus on a different set of features than a platform or tool dedicated to backups.

For a slightly deeper look at the differences between archives and backups and their individual goals, check out this short article from Iron Mountain (a company that does both archiving and backups for businesses).

High-Level Habits

Getting into a routine will help make sure the important things stay safe without having to spend a lot of time thinking about the process. Fortunately, a good backup system is largely set-it-and-forget-it, as long as it’s combined with well-organized data and the general habits we discussed last week. Here are three backup-specific habits to focus on:

1. Practice classifying all new data as “essential” or “nonessential” as soon as you encounter it.
   Essential data must be backed up. Nonessential data must not be backed up. It’s that simple. Nonessential data might not be useless or irrelevant, but it just doesn’t warrant the same kind of care, because you don’t need it that much or you could easily get it again if you had to. This is the safeguard-specific application of the new data habit #2 from last week, “Organize On Arrival.” Putting important data immediately into a location that is already configured to be backed up means you don’t have to think about it anymore.
   • Essential examples: tax records, personal letters, and digitized art from your kids
   • Nonessential examples: internet download catch-all folder, operating system files, temporary files, anything easily recreated
2. Run backups regularly.
   A good backup system doesn’t help if the backups are infrequent. Most systems allow you to schedule automatic backups to run on a weekly, daily, or even hourly basis. Take advantage of this to make sure your working copy is never too far ahead of your backups.
3. Test backups regularly.
   Backups that can’t be used to restore data are even worse than no backups at all since they provide a false sense of safety. However, most backup systems also provide some way to verify that each new backup copy is complete, intact, and readable. Don’t skip this step, even if you have to do it manually.

So what’s involved in a good backup system? Read on!

The 3-2-1 Backup Rule

A long-standing practice in the computing world, especially among corporate users, is the “3-2-1” backup rule, which means:

• 3 separate copies of your data
  One copy is the “live” or “production” data that you work with on a regular basis, and the other two are identical backups. Two separate backups are used to provide a redundant failsafe and to streamline the recovery process in certain situations.
• 2 different types of storage media
  One type of storage media will almost always be a hard drive, so the second type is often magnetic tape or, depending on the size and frequency of the backups, writable CDs or DVDs. Using different media types helps ensure that weaknesses of one media type (such as heat or humidity damage) don’t apply to the entire backup set. Therefore, catastrophic problems that affect one copy might leave the other one intact.
• 1 copy stored offsite
  Keeping one of the three copies of your data (one of the two backups) in a different physical location helps ensure that something like a fire, flood, or break-in doesn’t result in a total loss. Restoring from offsite backups is slower than restoring from a local copy, but it’s infinitely better than having nothing left at all.

Why three copies? Isn’t that overkill? While the odds of two copies failing simultaneously are extremely low, the remaining two rules are practically impossible to follow with fewer than three copies.

Modifications for Personal Users

The 3-2-1 backup rule is straightforward, but you might wonder whether it’s necessary to follow it to the letter, especially if you don’t have business-specific requirements such as document retention regulations to follow. The short answer is no, you don’t. Some people are actually quite antagonistic towards the original 3-2-1 backup rule because of the way people continue to follow it blindly despite a changing digital landscape.

The 3-2-1 rule can be modified in a few different ways that still provide a robust and dependable solution. With high-quality all-in-one backup platforms like Network Attached Storage (NAS) devices from Synology or QNAP as well as the rise of cloud backup platforms such as Backblaze, Crashplan, and dozens of others, it’s easy to find a solution that follows the spirit of the original rule without adhering to the letter.

Following the “[COPIES]-[MEDIA]-[OFFSITE]” count pattern from the original rule, here are a few modifications:

• 3-1-1: Here you have 3 copies of your data, only one storage type (hard disk), and one copy offsite. The first copy is your working data, the second is on a separate hard disk, and the third is on a cloud backup platform. Even though cloud backup services use physical hard drives just like you do (so there’s only one media type), the odds of both your storage and their storage failing simultaneously are astronomically low.
• 3-1-2: A slight modification of the above, some cloud backup providers explicitly provide redundant copies of your data in geographically separate data centers. This is the same amount of work for you as 3-1-1 above, but it might be slightly more costly. The product offerings of any cloud backup provider should explain whether they offer this feature either free or at an extra cost. If you cloud backups, this is good to have.
• 3-2-2: Here you have a second local media type, which for personal users almost always means a dedicated backup device like a NAS. Technically, these still rely on hard disks internally, but they provide extra redundancy through multiple drives and special filesystem design. (This is the method that I use personally.)
• 3-2-3: The “ultimate” level of redundancy, this solution follows the 3-2-2 pattern with a device such as a NAS, but then backs up that device offsite as well, typically by having a second NAS configured to mirror the first one in an alternate location such as a second office or a family member’s home (for personal users). This approach can be an ideal method for technically-minded families, with each household using a NAS for their main backup while mirroring it securely to other family members’ NAS devices.

If all these jumbles of numbers cause your eyes to glaze over, remember that many people are currently following the “1-1-0” rule and are one failure away from data loss. Getting a single external hard drive and leveraging your computer’s built-in backup software will upgrade you to “2-1-0”, which is significantly better.

Trust, But Verify

It’s easy to forget that just because you made a backup doesn’t mean you’re automatically safe. Verifying is always a critical final step. Nothing is worse than discovering in an emergency that you can’t actually retrieve your data for some facepalm-inducing reason.

Any reasonably good backup solution provides a way to test your backup, either in its entirety or by letting you browse through its contents. If there is an automatic verify-after-backup option, make sure you enable it. If not, set aside a few minutes on a regular basis (weekly or monthly) to make sure you can access backed-up data as expected.

It’s Okay to Start Small

Once again, remember that even one backup copy of your most important data on one external hard drive is a good place to start if it’s all you can do. It’s always possible to grow into more redundant complex systems as needed. Whether you stick to the original 3-2-1 backup rule or modify it to meet your needs, it’s absolutely worth the effort.

Do One Thing

Think about the data you have now and how much you should be backed up in case of a disaster. If you don’t have any backups in place yet, take the first step with an external hard drive (you might even have one already!) and your computer’s built-in system to begin protecting your most important files–or look into a solution like Backblaze or Crashplan or any of their competitors.

As always, small steps toward the goal will eventually bring you there.