Every editor has a horror story. A drive clicks once, then nothing. A RAID array rebuilds for 36 hours and fails halfway through. A laptop gets stolen from a car outside a client meeting. The footage is gone, and the deadline is not.
The 3-2-1 rule exists because drives fail. Not if. When.
What 3-2-1 Actually Means
The rule is simple. Keep three copies of your data, on two different types of media, with one copy stored offsite.
Think of it like a film negative. In the analog days, you had the camera negative locked in a vault, a work print in the edit room, and a dupe negative stored somewhere else entirely. Three copies. Two formats. One far away from the others. The 3-2-1 rule is the same logic applied to hard drives.
Three copies means your working drive, a local backup, and a second backup somewhere else. If your project lives only on your edit drive and a single Time Machine disk sitting next to it, you have two copies in the same location. That is not a backup strategy. That is a coincidence waiting to expire.
Two media types means your copies should not all live on the same kind of storage. If everything sits on spinning hard drives on the same desk, a single power surge can take all of them. Mix it up. An SSD for your working drive, an HDD for local backup, and cloud or LTO tape for the offsite copy.
One offsite means physically somewhere else. A different building. A different city. Cloud storage counts. A drive at a friend's house counts. The point is that a fire, flood, or theft at your primary location does not also destroy your backup.
What This Looks Like in Practice
For most freelance editors, a realistic 3-2-1 setup looks like this:
- Copy 1: Your working SSD or RAID, where active projects live.
- Copy 2: A local backup drive, updated nightly with software like Carbon Copy Cloner, ChronoSync, or rsync.
- Copy 3: Cloud storage like Backblaze B2, Wasabi, or even a Dropbox archive folder set to online-only.
The local backup handles the most common failure: a single drive dying. The offsite copy handles the catastrophic scenario. Together they cover nearly every realistic threat to your footage.
For larger shops with shared storage, the principle scales the same way. Your NAS is copy one. A second NAS or DAS is copy two. Cloud or a rotated set of LTO tapes is copy three.
The Part Most Editors Skip
The offsite copy is where most people fall off. Local backups are easy because the drive is right there. Offsite requires either paying a monthly bill for cloud storage or physically moving a drive to another location on a schedule.
Cloud makes this easier than it used to be. Backblaze B2 runs about six dollars per terabyte per month for storage alone. For a 10 TB archive, that is sixty dollars a month. Not trivial, but far cheaper than reshooting a project or losing a client.
The harder question is what actually needs to be in that offsite copy. Raw camera originals, yes. Final exports, yes. But what about the 4 TB of transcodes, proxies, and render cache files sitting in your project folders? Those can be regenerated. They do not need to be backed up, and they definitely do not need to be paying rent in the cloud every month.
Before You Back Up Everything, Know What You Have
This is where most backup strategies waste money. Editors back up entire project folders without checking what is actually inside them. Intermediate renders, unused media, orphaned clips from three revisions ago. It all gets copied, uploaded, and stored at full price.
Before you build your 3-2-1 system, it is worth knowing how much of your current storage is actually project-critical media versus generated files you could recreate in an afternoon. Clip Sweeper analyzes your projects and shows you exactly which files are in use and which are just taking up space. Cleaning house before backing up means your offsite copy is smaller, cheaper, and faster to restore when you actually need it.