What Is Redirect Chain? Definition and Examples

A redirect chain happens when the first redirect does not land on the final page but on another redirect, and so on. Each hop costs a network round trip for visitors and a fetch for crawlers. Chains build up naturally as sites go through rebrands, HTTPS moves and URL restructures, and they are easy to fix once found: point every old URL straight at the final destination.

How chains form

Chains are rarely designed. A site moves from HTTP to HTTPS, which adds one hop. Later it standardizes on www, which adds another. Then a page is renamed, and its old redirect still points to the previous name. A visitor following a years-old backlink can now pass through three or four redirects before seeing content.

A redirect loop is the broken version: A redirects to B, and B redirects back to A. Browsers stop with an error and crawlers give up.

http://example.com/features/sso
  -> 301 https://example.com/features/sso
  -> 301 https://www.example.com/features/sso
  -> 301 https://www.example.com/security/sso
  -> 200 OK

What Google does with chains

Google's documentation says Googlebot follows up to 10 redirect hops for a URL; beyond that, Search Console reports a redirect error and the content is not fetched. Short chains do work, but each hop is an extra request, can delay discovery of the final URL, and adds latency for real visitors, especially on mobile connections.

For robots.txt, the rules are stricter. RFC 9309 asks crawlers to follow at least five redirect hops for the robots.txt file itself, so do not chain your robots.txt.

Finding and fixing chains

Run a crawler over the site and over a list of old URLs (from old sitemaps, analytics, and your backlink tool). Most desktop crawlers have a redirect chains report. For a single URL, curl with the -L and -I flags shows each hop.

Fix chains by updating every redirect rule to point to the final destination, and update internal links so they skip redirects entirely.

Collapse each chain so the first hop goes straight to the final 200 URL
Handle protocol and host normalization in one rule where possible
Update internal links, canonicals, hreflang and sitemaps to final URLs
Re-crawl to confirm no loops or new chains were introduced

Example

Three rebrands, one chain

A fictional analytics company, Metricbird, started as metricbird.io, moved to getmetricbird.com, then to metricbird.com, and restructured its feature URLs along the way. Links from early press coverage now pass through four hops.

The fix is a single redirect map keyed on every historical URL pattern, each pointing at the current page, deployed at the edge so all legacy domains resolve in one hop.

Common mistakes

Only fixing the newest redirects

Old domains and protocol rules are where the longest chains start. Include every legacy hostname in the audit.

Leaving internal links on redirected URLs

Even a single hop is wasteful when you control the link. Point nav, footer and in-content links at final URLs.

Questions

One is ideal. Google follows up to 10 hops, but every extra hop adds delay and another chance of failure. Aim to resolve every legacy URL in a single redirect.

Google has said redirects do not lose PageRank, but a long chain can delay consolidation or fail entirely past 10 hops. Collapsing chains removes the risk.

Behavior varies by crawler and is not always documented. Shorter, direct redirects are safer for any client.

Sources

Redirect Chain