Broken Links SEO

A broken link is a hyperlink whose destination no longer exists or is unreachable. Click it and you hit a 404 or 410 error page. For search engines, every broken link is a wasted crawl request — they could have spent that time discovering or recrawling a page that actually matters. On a site with 10,000 indexed pages and 200 broken links, that is 200 requests per crawl cycle that return nothing useful. Users feel it too: a single broken link in your main navigation can raise your bounce rate by 15–20% on that session, I have seen it happen.

The fix is rarely complicated. You either restore the page, set up a [301 redirect](/301-redirect/) to the closest live equivalent, or leave it as a 410 if the content is gone for good (old products, seasonal offers). The mistake people make is treating all 404s as equal. They are not. A broken link on a page that gets 10,000 visits a month is a different problem from a link in an archived PDF that nobody reads. I prioritise by traffic, not by count. If you have 500 broken links but 480 of them are on low-traffic pages, you fix the 20 on your homepage and category pages first.

This is not about perfection. It is about protecting the crawl budget and the trust that users (and Google) place in your site. A [technical SEO](/technical-seo/) audit that ignores broken link severity is a list of problems, not a plan.

Broken links matter most right after a site migration, redesign, or content deletion. These are the moments when URLs change or disappear. If you have not mapped your old URLs to new ones with 301 redirects, every existing link to that old URL — internal or external — becomes a dead end. I once audited a site that had moved from http to https and forgot to redirect its blog section. They had 340 broken links overnight. Fixing them recovered about 12% of organic traffic within two weeks.

The second place broken links really hurt is in navigation and footer. These links appear on every page. One broken link in a top-menu item means every user on every page gets a potential error. That is a site-wide trust issue. I have seen a broken "Contact" link cause a 23% drop in contact form submissions over a month. Users do not always report it — they just leave.

Third, high-traffic pages. If your most visited product page links to a deleted user manual, that link is wasting authority and sending users to a dead end. I check these pages manually after any content update. The cost of a broken link scales with the traffic of the page that contains it. A simple rule: fix links on your top 20 pages before you touch the rest.

External links you control also matter. If you link to an external resource and it dies, your page looks neglected. I run a quarterly check on outbound links using a crawler, and I redirect any that I can control. For links I cannot control, I sometimes update the anchor text to say "originally linked to X (now offline)" so the user is not surprised.

Large sites (10k+ pages) need automation. Relying on manual checks is naive. I set up a weekly crawl that flags new 404s and sends them into a spreadsheet sorted by the traffic of the referring page. That way I only look at the ones that matter. If you have a site with 50,000 pages, you cannot treat each 404 as an incident. You treat the ones that appear in top paths.

Finally, [website structure](/website-structure/) changes often introduce broken links. If you restructure your category tree, every old category URL needs a redirect. I have learned the hard way: always run a pre- and post-migration crawl and compare the two lists.

I used to treat every 404 as an emergency. I would spend hours fixing broken links on low-traffic blog comments while ignoring the footer link that was broken on every product page. That was wasted effort. Now I know that some 404s are intentional — expired promotions, old product variants, seasonal pages — and they should return a 410 Gone to signal permanent removal. Chasing every 404 is like trying to fix every crack in the pavement when you have a hole in the roof.

My biggest mistake was deleting or renaming URLs without setting up redirects first. I once renamed a key category from /widgets/ to /widgets-guide/ because I thought the slug was better for SEO. I forgot to redirect. My organic traffic to that section dropped by 40% over the next month. That was a rookie move. Now I always map my redirects before I touch a URL, and I test them after deployment.

I also focused only on internal links in body content and completely ignored navigation, footer, and image links. I assumed users would not click a broken image. They do. A missing product image that breaks on a category page can kill the user's confidence. I now run a full-site crawl that includes all resources — CSS, JS, images — and I check for 404s on those too.

Another thing I got wrong: I thought external broken links did not matter much. "It is their site, not mine." Then I saw a client's page that linked to a dead industry report. The page was authoritative, but users who clicked the link saw a 404 and never came back. The bounce rate on that page was 10% higher than similar pages. I now check outbound links quarterly and replace or remove dead ones.

Finally, I underestimated how often broken links appear during a [site audit](/seo-audit/) because of relative URL issues or [canonical tag](/canonical-tag/) misconfigurations. A canonical tag pointing to a page that does not exist can create a broken link in the index. I now check canonical targets as part of the crawl, not as a separate step.