Sync Is Not Backup

Your sync tool is working perfectly. That's the problem.

01A mirror moves everything — including your mistakes

The appeal of sync is obvious. You delete a file on your laptop, and it disappears from your desktop. You save a document in one place, and it appears in another. The folder stays consistent across every device, and that consistency feels like safety. It isn't.

Sync is replication, not backup. The difference matters enormously. A backup preserves a point in time that you can return to; a sync tool propagates the current state of a folder to every connected copy as faithfully and as fast as it can. When you accidentally delete a project folder, the sync engine doesn't pause to ask whether you meant it. Within seconds — sometimes faster — that deletion is pushed to every device and every cloud copy attached to your account. You now have zero copies instead of several, and all of them are in perfect agreement.

This is the engine working exactly as designed. The problem was never the sync tool. The problem is calling it a backup when it isn't one.

02One copy with several windows is still one copy

Think of sync as a single logical drive projected across multiple screens. Your laptop, your desktop, your phone and the cloud storage endpoint are all displaying the same underlying data. Add ransomware to that picture: the malware encrypts the files on one device, the sync engine faithfully propagates the encrypted versions everywhere, and by the time you notice, every window is showing the same ciphertext. The "copies" weren't copies at all — they were mirrors.

The same logic applies to quieter disasters. A slow corruption that goes unnoticed for a week gets pushed to every device and overwrites any earlier version on each one. A misfired bulk rename propagates before you've reached for the keyboard shortcut to undo it. Even human error scales perfectly with sync.

Some sync services include a version history — a limited window during which you can retrieve a previous state of a file. This is genuinely useful, but it is not the same as a backup, for two reasons. First, the retention window is almost always short: a few days to thirty days on most standard plans, and only for file edits, not necessarily for deletions that get pushed as a confirmed intent. Second, version history lives inside the same service: if your account is compromised, locked or suspended, that history may be inaccessible at exactly the moment you need it.

A real backup lives somewhere independent. It cannot be overwritten by a sync event, corrupted by a bad client update, or deleted by whatever process is deleting your originals. It has versions and retention that you have chosen deliberately, stretching back far enough to catch the kind of slow, quiet damage that only becomes visible weeks later. And it has been tested, not just assumed.

The same four events, through a mirror and through a backup

When this happens	Sync	Backup
The laptop is stolen	Files come back	Files come back
You delete the wrong folder	Deleted everywhere; the trash may hold it briefly	Restore from any retained point
A file silently corrupts	The corrupt version is propagated	An older good version survives
Ransomware encrypts the folder	Encrypted copies replicate on schedule	An offline or immutable copy is untouched

This is genuinely useful, but it is not the same as a backup, for two reasons.

03What a real backup adds that sync cannot

The minimum meaningful backup has three properties sync doesn't provide. First, it is independent — written to a medium or service that isn't in the same propagation chain as your live data. An external drive that is connected only during backup windows qualifies; a second cloud folder in the same account doesn't. Second, it is versioned — it preserves multiple historical states, not just the latest one, so that you can go back past the bad moment. Third, it is retained — those versions are kept long enough to outlast the gap between when damage occurs and when you notice it.

The principle sometimes called 3-2-1 — three copies, across two types of media, with one stored off-site — is a practical expression of these properties. It sounds like overhead until the day sync does exactly what it's supposed to do.

Sync tools are not going away, and they solve a real problem: keeping working files consistent across devices. Use them for that. But the moment you find yourself thinking "I have two copies because I have sync," run the thought to its conclusion. Those two copies will respond identically to any event that touches one of them. That is exactly what you asked sync to do. It is not what you need a backup to do.

Sync is fast, automatic and extremely good at its job. Backup is slower, deliberate and preserves the past. You need both — and only one of them will be there when the other fails.

Most sync services do keep some history — a trash, a version list, a retention window. It is real and useful, and it is a setting somebody else controls.