How to Find and Fix Broken Links

A broken link is a link that no longer leads to a working page. Someone deleted the target, renamed its URL, or the whole domain went away. Visitors land on an error page, and crawlers spend part of their visit on a dead end.

The fastest way to find them is to check one page at a time. Run the free Siteiz Broken Link Checker on a page and you get every link on it, each with a live status.

Step 1: Check the pages that matter most

You do not need to crawl a whole site to find most broken links. Most sites reuse a handful of templates, so the same broken link tends to repeat across hundreds of pages. Check these first:

1. Your homepage
2. One category or listing page
3. One product or service page
4. One blog article
5. Any resource page that links out a lot

A broken link in the header or footer shows up on all of them. Fix it once in the template and it is fixed everywhere.

Step 2: Read the status, not just the colour

Not every red result means the same thing. Here is what each status tells you and what to do about it.

Status	What it means	What to do
404 or 410	The page does not exist	Point the link to the right page, or remove it
5xx	The server failed	Re-check later. Remove the link if it keeps failing
No response	Domain gone, refused, or timed out	Check the domain in a browser. Remove if it is dead
Redirect chain	Two or more hops before the final page	Link straight to the final URL
401, 403, 429	The site blocked the checker	Open it in a browser before you change anything

That last row matters. Many sites refuse automated requests even when the page works fine for people. A checker that calls those links broken is guessing. The Siteiz checker labels them Blocked and leaves the decision to you.

Step 3: Fix internal links at the source

For a broken internal link, the right fix is almost always to update the link itself, not to add a redirect. A redirect works, but it adds a hop for every visitor and every crawler, forever.

If an old URL has backlinks from other sites, keep a single 301 redirect from the old URL to the new one, and still update your own links to point directly at the new URL.

Step 4: Clean up redirect chains

Chains build up over time. A page moves from http to https, then gets a trailing slash, then a new slug. Each change was fine on its own, and together they add up to three hops. Link to the final URL and remove the middle steps where you can.

The link problem most checkers miss

There is a second way a link can fail, and a status code will not show it. If your navigation, related-posts block, or footer is built with JavaScript, the links may be missing from the HTML your server sends. They only appear after the browser runs the scripts.

People can click those links. Googlebot renders JavaScript, so it usually finds them too. AI crawlers such as GPTBot, ClaudeBot, and PerplexityBot read the raw HTML and do not run scripts. For them, the link does not exist, and neither does the path to the page behind it. We explain why in Does ChatGPT read JavaScript?.

The Siteiz checker compares the raw HTML with the rendered page and flags these links as JS-only. The fix is to render key links on the server as plain <a href> elements.

A simple routine

• Check your main templates once a month, and after every release that touches URLs.
• Fix internal 404s the same day. They are the cheapest wins you have.
• Re-check server errors before you remove anything.
• Treat JS-only navigation as a real issue if you care about AI search.

Working links are the floor, not the ceiling. To see how AI crawlers read the rest of the page, run the free AI visibility scan.