The worst support tickets start the same way: 'I updated/deleted/changed something and now...' — with no backup. The two-minute ritual prevents every one of them.
The ritual, by change type **Plugin/theme updates (WordPress):** 1. UpdraftPlus → 'Backup Now' (database + files) 2. Update ONE plugin at a time 3. Check the site after each — the breakage, if any, is identified by name
**Code edits (any platform):**
1. File manager → download the file (or rely on 24-hour versioning — Kairalmas keeps it automatically)
2. Edit, save, verify in a private window
**Database changes (imports, big edits):**
1. phpMyAdmin → Export → save the .sql locally
2. Then and only then, import/change
**Migrations/redesigns:**
1. Full backup: files + database + a laptop copy
2. The migration checklist's step 1 exists for a reason
Why version history is not enough alone Kairalmas's 24-hour file versioning covers file oopses beautifully. But database changes live OUTSIDE files — the UpdraftPlus/PHPMyAdmin backup is the only thing standing between a bad SQL import and a rebuilt site.
The psychology (why the ritual fails) The change takes two minutes; the backup feels like overhead — until the day it feels like salvation. Reframe: the backup IS part of the change. Not backing up before an update is like unbuckling because the drive is short.
Two minutes. Every time. The habit costs less than one lost evening — and this article is written from a support inbox that has seen the difference.