Skip to content
Rescue 404

Repair

WordPress Update Broke My Website — How to Fix It Step by Step

Calm recovery steps when a WordPress core, plugin, or theme update breaks your site — what to document, how to regain access, when to roll back, and when to call emergency repair.

Josh

Abstract illustration of a website dashboard cracked by a failed update with a recovery path highlighted
After a bad update: document → regain access → isolate the change → restore or roll back → verify critical paths — then harden with Care.

You updated WordPress — or WordPress updated itself — and the site is gone: white screen, critical error, 500 response, broken layout, or a checkout that never finishes. This is one of the most searched emergency queries for a reason. Owners need a recovery path, not a lecture about “always test on staging” after the fire starts.

Here is a practical sequence that matches how emergency website repair actually begins: stabilize evidence, regain access, isolate the change, restore or roll back, then verify.

What usually broke

Most post-update failures cluster into a few patterns:

  • Fatal PHP errors after core, plugin, or host PHP changes
  • Plugin/theme conflicts that only appear on certain templates or admin screens
  • Memory exhaustion during or after background updates on constrained hosting
  • Partial updates that left files half-written when the process timed out
  • Builder or cache collisions (Elementor, Divi, Rocket, etc.) after dependency bumps

The trigger is often obvious (“Update All Plugins”). Sometimes it is invisible (auto-update overnight). Treat the timeline as evidence.

Step 1 — Document before you touch more things

Capture:

  • Exact URLs that fail (homepage, /wp-admin/, checkout, one post)
  • Full error text or HTTP code
  • Time noticed (with time zone)
  • What changed: core, plugins, theme, PHP version, host ticket, DNS/SSL edits
  • Whether the public site, admin, or both are broken

Screenshots help. So does not making a broken site worse with random “fix” plugins.

Step 2 — Confirm access paths

Try, in order of least damage:

  1. Front end in a private window (rules out your cache/cookies)
  2. /wp-admin/ and /wp-login.php
  3. Hosting panel / file manager / SFTP
  4. A known backup or host snapshot tool

If admin is dead but hosting works, you can often disable the last updated plugin via file rename. If hosting is unreachable, recovery belongs with the host and a technician — not more dashboard clicks.

Prefer a dedicated repair admin account later so activity stays attributable. Never paste production passwords into chat or intake forms; use a one-time secret link after you are in contact.

Step 3 — Isolate the last change (safely)

When you have file or panel access:

  • Rename the most recently updated plugin folder (adds .disabled or similar) and retest
  • If a theme update is the suspect, temporarily switch to a default theme only if you understand the front-end impact and have a rollback plan
  • Clear opcode/page cache after each change so you are not staring at a cached failure
  • Check debug.log or host error logs for the first fatal line — that name is usually the culprit

Do one change at a time. Re-enable only after the site boots.

Step 4 — Restore or roll back when isolation stalls

If the site stays down, revenue is offline, or you lack confidence:

  • Restore the newest known-good backup to a safe target when possible
  • Verify the restore (backup vs restore verification) — login, forms, checkout, SSL
  • Only then decide whether to retry updates one-by-one on staging

A restore that works beats a heroic three-hour debug while phones ring unanswered.

Step 5 — Verify the business paths

After the site loads again, check the paths that pay bills:

  • Homepage and primary service pages
  • Contact forms and notification email
  • WooCommerce cart/checkout and payment if you sell online
  • Admin login and ability to edit content
  • SSL padlock and mixed-content warnings

Then pause automatic “update everything” until you know what failed and why.

How Care prevents the next break

DIY update night is how many small businesses meet emergency repair. A Website Care Plan exists so updates are supervised, backups are restore-tested, monitoring catches downtime, and someone can act during Eastern Time hours when a patch goes wrong.

You do not need Care to hire a rescue. Repair first if the site is down. Care is how you stop living on update roulette.

When to call emergency repair now

Stop DIY and request emergency repair if:

  • The site or checkout is down and you cannot regain admin/hosting access
  • You have no verified backup
  • You see malware symptoms mixed with the update failure
  • You are about to delete wp-content “to start fresh”

Bring the URL, timeline, hosting login path, and any error text. That is enough to start a paid diagnostic and a written repair quote.

Frequently asked questions

Why did a WordPress update break my site? +
Updates change PHP assumptions, APIs, and asset loading. A plugin or theme that worked on the old version can fatally error, loop redirects, or exhaust memory after the change — especially on shared hosting with tight limits.
Should I keep clicking Update on more plugins to fix it? +
No. More updates during an outage often widen the blast radius. Document the failure, regain stable access, then isolate one change at a time.
Is restoring a backup safer than debugging live? +
Often yes when revenue is offline and you have a recent, known-good restore point. Prefer restoring to a safe target or rolling back the last change rather than editing production blindly.
Can automatic updates break a site even if I did nothing? +
Yes. Background core or plugin updates can collide with themes, page builders, or low PHP memory. That is why supervised updates and staging matter on business sites.
When should I stop DIY and hire repair? +
Stop if you cannot reach admin or hosting, if checkout or forms are down, if you lack a verified backup, or if you are about to delete folders you do not fully understand. Emergency repair exists for that hour.