Redirect chains and loops: find them and flatten them
A redirect chain is what you get when one redirect leads to another. You ask for page A, the server sends you to B, and B sends you on to C before you finally land somewhere. A redirect loop is the broken cousin: A points to B and B points back to A, so the browser circles until it gives up and shows an error.
Both creep in slowly. You move a page, then move it again a year later and forget to update the first redirect. You add an HTTPS rule on top of a www rule and they start fighting. None of it is dramatic on its own, but chains and loops quietly cost you crawl budget, speed, and a little ranking value at every hop. This guide covers why they hurt and how to flatten them.
How chains and loops happen
Almost nobody sets out to build a redirect chain. They accumulate. A page moves from one URL to another, and you set the redirect. Later the page moves again, and you redirect the second URL to the third, but the original redirect still points at the now-defunct second URL. Now anyone arriving at the first URL takes two hops to reach the live page.
Loops usually come from rules that conflict. A common one: a rule forces every request to HTTPS, and a separate rule forces a switch between www and non-www, and the two end up handing the request back and forth. The browser follows redirects up to a limit, then stops and shows a 'too many redirects' error. The page is effectively dead even though nothing returns a real error code on its own.
Why chains hurt crawl budget and speed
Every hop in a chain is a separate round trip: the browser or crawler makes a request, gets a redirect, and makes another request. For a visitor on a slow connection, two or three hops add real delay before the page even starts loading. For a search crawler, each redirected URL is one it has to fetch and process.
Crawlers work within a budget, a rough limit on how much of your site they will fetch in a given window. When that budget gets spent following chains instead of reaching real content, fewer of your actual pages get crawled and updated. On a large site this matters: chains can leave new or changed pages waiting longer to be reindexed.
Why chains leak ranking value
A 301 passes nearly all of a page's ranking strength to its destination, but not quite all. The loss at a single hop is small enough to ignore. Stack three or four hops in a row and those small losses compound, so the final page inherits noticeably less than it would from a direct redirect.
There is also a simpler risk. The longer the chain, the more chances for one link in it to break. If the middle URL in a three-hop chain gets cleaned up or starts returning a 404, everything downstream of it stops working and the ranking strength that was flowing through evaporates. A single-hop redirect has no fragile middle to break.
How to find them
You cannot see a chain by clicking a link, because the browser quietly follows the whole thing and only shows you the final page. You have to inspect the redirect path directly. A redirect checker requests the URL and reports each hop and its status code, so a chain shows up as a list of two or more redirects before the final 200.
For a whole site, crawl it and look for any URL whose path includes more than one redirect, plus anything flagged as a redirect loop or a 'too many redirects' error. Pay attention to old high-traffic URLs and anything other sites link to, since those are the chains that cost you the most.
How to flatten them
Flattening a chain means rewriting each redirect so it points straight at the final live URL. If A goes to B goes to C, change A so it goes directly to C, and retire the hop through B if nothing else needs it. After the change, every old URL reaches its destination in one jump.
For loops, find the conflicting rules and decide on a single canonical form for your URLs, then make sure your redirects all push toward that one form rather than fighting over it. Pick HTTPS plus your preferred www choice, write the rules so they cooperate, and test the URL again to confirm it now resolves in one hop instead of circling.
Frequently asked questions
- How many redirects in a row is too many?
- The ideal is one. A single 301 from the old URL straight to the live page is what you want. Two hops is worth flattening, and three or more should be treated as a problem. Browsers and crawlers also enforce a hard limit, after which they abandon the request entirely.
- Do redirect chains actually affect SEO?
- Yes, in two ways. They waste crawl budget, so fewer of your real pages get crawled and reindexed, and they leak a little ranking value at each hop. Neither is catastrophic on its own, but on a large or heavily linked site the cumulative effect is worth eliminating.
- What causes a redirect loop?
- Conflicting rules, most often. An HTTPS redirect and a www redirect that disagree on the canonical URL can hand the request back and forth until the browser gives up. Decide on one canonical form, then write the rules so they all push toward it instead of fighting.
- How do I see the full redirect path for a URL?
- Use a redirect checker, which requests the URL and shows each hop and its status code rather than just the final page. A clean URL shows one 301 then a 200. A chain shows several redirects in a row, and a loop shows the same URLs repeating or a too-many-redirects error.
Related tools
Redirect Checker
Trace the full redirect path for any URL, hop by hop, and spot chains and loops at a glance.
301 Moved Permanently
Reference for the permanent redirect that should sit at the end of a flattened chain.
308 Permanent Redirect
Reference for the permanent redirect that preserves the request method, the method-safe sibling of 301.
Related guides
Put this guide to work on your own site
Sitewell’s free tools run the exact checks covered above — no signup, results in seconds.