301 redirects: how to avoid mistakes that leak SEO value
Redirects should carry your rankings forward when a page moves. Here is how 301s, chains, and loops work, and the mistakes that quietly drain their value.
Every site moves pages eventually: a URL structure changes, two articles get merged, a product goes out of stock. A redirect is supposed to carry the old page's ranking signals to the new one. Done wrong, it does the opposite. It wastes crawl budget, splits signals across multiple URLs, and can leave both search engines and AI crawlers unsure which page actually answers the query.
How a 301 redirect actually works
A 301 tells a crawler that a URL has moved permanently, and that the destination should take over its ranking signals. Most search engines consolidate the old page's links and authority into the new one, though the transfer is rarely perfect and takes time. A 302 says the move is temporary, so the crawler keeps the original URL as the canonical one and does not pass signals the same way. Meta refresh tags and JavaScript-based redirects work eventually for many crawlers, but they add delay and are less reliable for AI crawlers that often fetch a page once and move on without executing scripts.
The practical rule: use a 301 for anything permanent, a 302 only for anything genuinely temporary, and never leave a 302 in place for months. If it is still there after a quarter, it is not temporary anymore, and search engines start to treat it that way regardless of what the status code says.
Redirect chains and loops
A redirect chain happens when URL A redirects to B, which redirects to C. Each hop adds latency, and each hop is a chance for a crawler to give up before reaching the final page. Google has said it will follow a handful of hops, but it will not follow them indefinitely, and AI crawlers tend to have even less patience since they are usually fetching a page to answer one specific query, not to index a whole site.
Chains build up quietly. A site redesign creates one redirect, then a later migration adds another on top instead of updating the first one to point straight at the final URL. After a few years, sites often carry chains three or four hops long without anyone noticing, because a browser follows them instantly and nobody checks the network tab.
A redirect loop, where A points to B and B points back to A, is worse: it returns an error to both crawlers and users instead of a page. Loops usually come from conflicting rules in different layers, such as a CDN redirect and an application-level redirect that disagree with each other.
Mistakes that quietly leak ranking value
- Redirecting every dead page to the homepage instead of the closest relevant page. This tells search engines the content had no real replacement, which is worse for rankings than a clean 404.
- Leaving internal links pointing at the old URL after adding a redirect. Every internal link that still passes through a redirect wastes a hop and delays the crawler.
- Using a 302 for a permanent move, often because it is the default in a CMS or framework and nobody changes it.
- Redirecting HTTP to HTTPS and www to non-www (or the reverse) as separate hops instead of one, which doubles the chain for every single URL on the site.
- Forgetting to update the XML sitemap, so it keeps listing pre-redirect URLs long after the move.
- Redirecting a page to a URL that later gets deleted too, creating a chain that dead-ends without anyone noticing until traffic drops.
How to audit your redirects
- Crawl the site with a tool that reports redirect chains and loops, not just individual redirects.
- Pull the list of URLs currently returning a 301 or 302 and check each one resolves in a single hop.
- Find internal links that point at a redirected URL, and update them to link straight to the destination.
- Check the XML sitemap only lists live, 200-status URLs.
- Review any 302s older than a few months and decide if they should become 301s or be removed.
- Spot-check that redirected pages land on genuinely relevant content, not a generic homepage or category page.
None of this is glamorous work, and it rarely shows up on a roadmap next to a new feature. But a site that has been through two or three redesigns without a cleanup usually has redirect debt that is costing it crawl efficiency and diluting the very signals a migration was supposed to protect. A single, direct redirect from an old URL to its true replacement is worth more than a clever chain that eventually gets there.
A redirect that takes three hops to reach its destination is asking every crawler, human and AI alike, to trust you three times instead of once.
Next time you plan a URL change, write down the final destination first, then point every redirect straight at it. Fix the internal links in the same pass. That one habit prevents most of the redirect debt sites accumulate over years.