Most hosts back up your site once a day, in production, and hope nothing goes wrong. We snapshot at the storage layer every 4 hours, replicate off-site, and let you roll back any file or table in under a minute, with zero impact on your live site.
Every plan includes snapshots every 4 hours with 30 days of retention. If your store, membership site, or publication updates constantly, hourly snapshots shrink your worst-case data loss from 4 hours to 1 hour. Add it during checkout or from the control panel at any time.
Every hour, 24/7
720 restore points per 30 days instead of 180.
Same 30-day retention
Hourly snapshots are kept for the same 30 days, no longer.
Once-a-day, taken at peak traffic, stored on the same disk, with a restore that requires a support ticket. We did the opposite of all four.
If something goes wrong at 3pm, the most recent backup is from 2am. Every order, comment, form submission and edit since then, gone. With snapshots every 4 hours, your worst case is 4 hours, not a full day. If you want it tighter, hourly snapshots are a $5/mo per-account add-on.
We snapshot at the storage layer (ZFS), not the application. Your CPU, RAM and database aren't touched. Visitors don't notice; you don't get a 'site is sluggish' ticket from a customer mid-backup.
Backups stored on the same machine as the site help you recover from a bad plugin, not from a hardware failure. We replicate every snapshot to a different region on a different provider, so the worst day still has an undo button.
A pipeline designed so the only thing you ever have to do is pick a point in time.
01
ZFS storage takes a copy-on-write snapshot of your site's filesystem and database. The operation is atomic and finishes in milliseconds, your site keeps serving traffic without a hiccup.
02
Snapshots live on dedicated backup storage, not on your site's NVMe drive. A bad disk on the live server doesn't take your history with it.
03
Every snapshot is replicated to a second region within minutes. The copies are encrypted at rest with per-customer keys.
04
You always have around 180 restore points (6 per day, 30 days) to choose from. Pick a moment, preview, restore, files only, database only, or the whole site.
Plenty of hosts will sell you a backup. Far fewer will let you actually use it without opening a ticket and waiting four hours. We obsessed over the restore flow, because that's the part that matters.
See it in actionStep 01
Browse 30 days of 4-hour snapshots. Filter by date, search by event (we tag deploys, plugin updates and core upgrades).
Step 02
Full site, files only, database only, or a single file/table. Useful when a plugin update broke one page but you don't want to roll back yesterday's orders.
Step 03
Spin up a temporary clone of the snapshot on a staging URL. Verify, then promote, or discard and try a different point in time.
Step 04
Median restore in under a minute for sites under 5 GB. Your live site stays online during the restore, we cut over atomically when the new copy is ready.
Real situations our team handles every week, and the difference between 4-hour snapshots and daily is the whole story.
With daily backups, you'd lose the morning's orders rolling back to 2am. With 4-hour snapshots + per-table restore, we roll back just the plugin's database tables to noon, and every order taken in between is preserved.
Browse to the most recent snapshot, restore that single file/post, and you're done. No full-site rollback. No lost work elsewhere.
We pinpoint the exact snapshot before the injection, restore the affected files, rotate the keys, and have you clean again, usually in under 10 minutes.
Everything teams ask before trusting us with their site's history.
Still have a question? Ask our team
