Links are how both visitors and crawlers move through a site. When they break, people leave and crawlers spend their visit on pages that no longer exist.
Find them first
Broken links come in two kinds, and they need different searches:
- Internal: your pages linking to your own dead pages. A crawl of your site finds all of them.
- External: other sites linking to pages you deleted or renamed. These are valuable, because someone chose to link to you. Search Console shows 404s that had visitors, and your server logs show the rest.
The right fix for each case
The page moved. Redirect the old address to the new one with a 301. Do it even if you also updated your own links, because other sites and old bookmarks still use the old address.
The page is gone for good and has no replacement. Let it return 404 (or 410), and remove your own links to it. Redirecting everything to the homepage is worse than a 404: visitors land somewhere irrelevant, and search engines treat these "soft 404s" as errors anyway.
The page is gone but something similar exists. Redirect to the closest real replacement. A discontinued product goes to its category or its successor, not to the homepage.
The link has a typo. Fix the link. No redirect needed.
Redirect chains
A chain is a link that goes to a page that redirects to another page that redirects again:
/old-post -> /blog/old-post -> https://site.com/blog/old-post -> /blog/new-title
Every hop costs a request. Some crawlers give up after a few, and each hop is a chance for a mistake to creep in.
Chains build up naturally: one redirect when the URL structure changed, another when the site moved to HTTPS, a third when a post was renamed. Nobody adds a chain on purpose.
The fix: point the first redirect straight at the final address, and update the links on your own pages to skip the redirect entirely.
Redirect loops
A loop is a chain that returns to where it started. Browsers show "too many redirects" and the page is unreachable. The usual cause is two rules that disagree: one forcing www, another removing it, or an HTTPS rule and a language rule fighting each other. Read the rules in order and make sure only one owns each decision.
Keep them from coming back
- When you rename a page, add the redirect in the same change, not later.
- Keep a single list of redirects in one place, and review it when the site structure changes.
- After any migration, crawl the whole site before you tell anyone it is live.
- Re-crawl monthly. External sites disappear and their links rot without you touching anything.
What Rankwell shows
The site audit reports every internal link that leads to an error page, with the page that contains the link, so you know where to edit. Redirect chains with more than one hop are listed with the full path, and the crawled-pages table marks which addresses redirected and where they landed.
For the pages other sites link to, check Search Console as well: those links are worth keeping alive.