Find and Fix Broken Links in WordPress Without Server Lag

Broken links quietly cost you visitors and trust. Here's how to scan a WordPress site one post at a time, so its size never causes a timeout, and fix them.

Broken links are invisible damage. Nobody reports them—a visitor just clicks, hits a 404, and quietly leaves. Meanwhile search engines crawl the same dead URLs and file your site under “not well maintained.” I’ve inherited sites with dozens of broken links the owner had no idea existed, because who re-checks a link they added two years ago? The catch is that finding them has a reputation for hammering your server. It doesn’t have to. Here’s the painless way.

The essentials

  • Broken links cost you visitors and trust—and they accumulate silently as external sites die and URLs change.
  • A checker that scans the whole site in one pass chokes on big sites; the Broken Links module scans post by post, one request per post from the admin tab you keep open, so the size of your site never causes a timeout.
  • It checks internal and external links with lightweight HEAD requests and a configurable timeout (10s by default).
  • Each dead link gives you four actions: recheck, edit the source post, dismiss a false positive, or unlink it (the anchor text stays).
  • Scans can run manually or on a daily/weekly schedule, with an email alert when a scheduled scan finds something new.

Broken links hurt a WordPress site three ways: a visitor who lands on a dead page leaves instead of reading on, a broken internal link cuts off a page nobody can reach any more, and a site full of dead links reads as neglected—to people and to search engines alike. And they accumulate on their own. Every dead link has a cost, and they stack up:

  • User experience — someone who lands on a 404 page rarely tries again; they bounce.
  • SEO — search engines keep requesting dead URLs instead of your live content, and a mass of broken links reads as a neglected site. (Google’s crawl budget advice only concerns sites of 10,000+ pages, so for most sites the cost is the lost visitor, not the crawl.)
  • Lost paths — a broken internal link is a page nobody can reach, however good it is.

And they appear on their own: an external site goes offline, you restructure your URLs, or you delete a page that other posts still link to. Left alone, a healthy site slowly rots at the edges.

Why does a whole-site scan slow your site down?

A link checker slows your site down when it scans everything in one pass: on a site with thousands of posts, loading every link into a single request either grinds the server to a crawl or hits the PHP execution time limit and dies half-finished, so you never get a complete report. On a ten-post blog, the same approach works.

The fix is architectural: check one post at a time. The Broken Links module sends a separate request per post from the admin page you keep open, so the work is spread into small, safe chunks the server barely notices. A 10,000-post site scans the same way a 10-post site does—just over more steps. It also uses HEAD requests (wp_remote_head), which ask a server only “does this exist?” instead of downloading the whole page, keeping each check fast and cheap. A single post full of unresponsive links can still take a while: each link waits up to the timeout you set.

To scan for broken links without overloading your server, use a checker that works one post at a time. In Blaminhor Essentials, activate the Broken Links module, start a scan from its admin page, and review each dead link by its HTTP status code—then recheck it, edit the source post, unlink it, or dismiss a false positive.

Google Search Console’s Page indexing report shows the dead URLs on your site that Googlebot ran into; it says nothing about the links in your content pointing elsewhere, which is what this scan covers. Step by step:

  1. Open the module and start the scan. It runs from that browser tab, one request per post, so keep the tab open until it finishes. Results stream in live as each post is processed, and you can stop at any point (a new scan starts over from the beginning).
  2. Review what it finds. Each broken link is listed with its HTTP status code—404 (gone), 403 (forbidden), a timeout, and so on—so you know why it failed.
  3. Act on each one:
    • Recheck — maybe the server had a hiccup; one click re-tests the URL.
    • Edit — jump straight to the source post to fix or remove the link.
    • Dismiss — mark a known-good link as a false positive so it stops nagging you.
    • Unlink — remove the dead link from the post but keep its text, when there’s nothing left to point it at.

The status code tells you why a link failed and what to do about it:

HTTP statusWhat it usually meansSensible action
404 Not FoundThe target page is gone—deleted or movedEdit the source post; add a 301 redirect if it moved
403 ForbiddenThe server refused both the HEAD check and a confirming GET—often a bot filterRecheck; dismiss if the link opens fine in a browser
Status 0 with a timeout messageNo response within the limit (10s by default); the module shows no HTTP code, just the error textRecheck—it may be a temporary hiccup
Status 0 with a connection errorThe host is offline or its domain is goneEdit or remove the link

Illustration of broken links: snapped chain links scattered between browser windows, with magnifying glasses, green checkmarks and red crosses. Illustration. In the module’s own list, each dead link appears with its status code and source post—recheck, fix at the source, or dismiss a false positive.

For a broken internal link where the page genuinely moved, the cleanest fix isn’t just editing the link—it’s a 301 redirect from the old URL, which also rescues any external links you don’t control. The scan looks outward, at links inside your content; for the opposite direction—visitors arriving on URLs that don’t exist on your site, including the ones AI assistants invent—see how to fix 404 errors from ChatGPT.

Can you set it and forget it?

Yes. Leave scanning manual, or set it to run daily or weekly on a schedule; turn on email notifications and a scheduled scan that discovers new broken links emails you once that pass finishes. Between scans, a dashboard signal keeps a running count of links still waiting on you, so you find out when a link dies, not months later.

A scheduled pass runs in the background in short slices, so it can take a while on a big site—the email arrives when the whole pass is done, and only counts links that weren’t already dismissed.

Broken links never fix themselves, but keeping on top of them is a five-minute habit, not a chore. Scan after any migration or URL change, after bulk-deleting content, and on a light recurring schedule the rest of the time—start, review, fix, done.

Broken Links is one of the 26 modules in Blaminhor Essentials—free and open-source on WordPress.org.

Download Blaminhor Essentials

– blaminhor

FAQ

Does the broken link checker slow down my visitors?

No. The scan runs in the admin on demand or on a schedule, not on front-end page loads, so visitors are never affected. Each link is checked with a HEAD request and a timeout, and the scanner deliberately refuses to probe private or reserved network addresses for safety.

How can I check for broken links without a plugin, using Search Console or online tools?

You can. Google Search Console's Page indexing report lists the URLs on your site that Googlebot found returning 404, and free online crawlers like Dr. Link Check scan your site from the outside. Both work, but they run occasionally and off-site—so a built-in scanner that rechecks on a schedule and links straight to the source post stays far more current.

Does it detect broken images too, or only links?

Only links. The Broken Links module extracts the `<a href>` links in your content—including links that point to a file, such as a PDF—and reports any that fail, so you can swap the reference or re-upload the file. An image embedded with an `<img>` tag isn't scanned, so a missing image file goes unnoticed unless something also links to it.

What's the difference between a broken link and a 404 error?

A broken link is any link that fails to reach a working page; a 404 is one specific reason why—the server replying "not found". A link can also break with a 403, a 500 or a timeout. That's why the checker shows the real HTTP status code, not just a blanket "broken".

Can broken link scans run automatically?

Yes. You can leave scanning manual or schedule it to run daily or weekly. With email notifications enabled, a scheduled scan that finds new broken links emails you so you can fix them without having to remember to check the report yourself.

How often should I check my site for broken links?

For most sites a weekly scan is plenty (the schedule offers daily or weekly); a busy blog with lots of outbound links benefits from daily. Because external sites die without warning, the safest habit is a scheduled scan with email alerts, so you hear about new dead links instead of relying on your memory to look.

blaminhor Building what's missing.

Comments