Blog / WordPress /

WooCommerce and WordPress: database backups before plugin changes

Plugin updates are boring. Right up until one changes how your orders, products or settings are stored.

That’s when you find out whether your backup is real. A backup only gives you a way back if it holds the data you need and you know how to restore it. “I think the host does backups” isn’t a plan.

Key takeaways

  • Back up both the database and files before a consequential change.
  • Check that the backup completed and can be restored.
  • Test plugin changes away from the live store when possible.

Know what your store depends on

WooCommerce product and order records live in the database, while themes, plugins, uploads and some configuration live in files. A database-only backup may preserve orders but miss a customised plugin or uploaded asset. A file-only copy may leave you without recent orders. Know what your backup tool captures and how often it runs. When I take over a WooCommerce site, that’s one of the first questions I ask the host.

Before updating a plugin, note its current version and the change you are making. Review the release notes and check whether it affects checkout, payments, shipping, product data or another critical workflow.

Verify the recovery path

A green “backup complete” message is nice. It doesn’t prove the restore will work. I don’t trust a backup I haven’t seen restored. If you can, restore a copy to a staging environment and check the front end, admin area and a sample order. Keep backups protected and separate from the live site so an account or server problem does not remove both.

For a live change, choose a quiet period and make sure someone can monitor the store. After updating, test a product page, cart, checkout and any integration that matters to your setup. If something breaks, follow the rollback plan rather than making several uncertain changes at once.

Write down where the backup is, when it was taken and who can restore it. Before the next plugin update, can you explain how you would get the store back to its previous state?

Make the backup useful before the change

A recovery copy is only useful if it matches the way the site stores information. On a WooCommerce shop, the database changes whenever orders arrive, products are edited or customers update details. The files may change less often, but the active theme, custom code, uploads and plugin configuration can still matter. Ask what the backup product includes rather than relying on a label such as “full”. If the answer is unclear, find its documentation or ask the host before you need a restore.

Timing matters as well. A copy made last night may not contain today’s orders. For a content site, that gap might be manageable; for a store, restoring an older database could remove transactions or changes made since the copy. Do not restore over production as a quick experiment. Use a staging copy, keep it protected, and confirm that the backup date and contents fit the task.

Plan the change, not just the copy

Write down the current plugin version, the version you intend to install and the reason for the update. Check whether the release notes mention a change that touches checkout, payment, shipping, stock, tax or a connection to another service. A small update can have a large operational effect if it sits in the middle of a workflow.

Choose a time when the person responsible can observe the site. Let the team know that checkout or admin behaviour is being checked. After the update, test the specific function that motivated the change and a few neighbouring journeys: product page to basket, basket to checkout, and the confirmation a customer should see. Do not place a real order unless the store’s test procedure allows it.

A useful change note is short: what changed, who made it, when, how it was checked and where the recovery copy is. If something behaves differently, that record helps distinguish a plugin issue from an unrelated change. If you need to roll back, check whether restoring the database could wipe newer orders. Bring in the right technical owner. Don’t press restore in a panic.

Keep a recovery routine current

Backups should be monitored, access-controlled and retained according to the business’s needs. A successful scheduled job is not the same as a tested recovery. A periodic restore test on a safe environment reveals missing files, credentials or steps while there is still time to fix them.

The level of testing should match the risk. A checkout extension deserves more care than a small wording adjustment. Make the difference explicit in the team’s change process, so every update is neither treated as harmless nor turned into an unnecessarily dramatic event. Before the next consequential plugin change, can you point to the backup, the restore owner and the exact checks you will run?

Record the result

After the check, note whether the expected workflow completed and where you looked for failure. A quick visual scan is not a substitute for testing the important path. For a shop, that may mean checking that a product can be added to the basket and that the checkout steps load as expected, using the store’s approved test process. If the update affects an integration, verify that connection too.

Keep the note factual. If something failed, capture the error and the time rather than immediately assigning blame to the plugin. That makes the next investigation quicker and gives the site owner a useful history of changes. If the update succeeds, the note still has value: it records what was checked and who can approve the next change.

Here’s my one ask: before your next plugin change, write down the recovery owner, the data at risk and the checks that prove the store still works. And be honest. When did you last actually restore one of your backups?