redirect chain

A single redirect is normal and safe. A chain forms when redirects stack: http:// to https://, then www to non-www, then an old slug to a renamed one. Each 3xx response is a separate round trip, so a three-hop chain makes the visitor (and Googlebot) wait on four requests instead of one.

Why the hops matter

Google’s John Mueller has said Googlebot follows “up to 5 hops in the chain per crawl attempt.” Past that, Googlebot stops for that crawl and picks the chain up on a later visit, which delays how fast the final URL gets crawled and indexed. Long chains also undercut the consolidation you wanted from the redirect in the first place: signals pass through more junctions, and any one misconfigured hop (a stray 302 in the middle, a loop, a typo) can break them.

The usual causes are protocol and host normalization piling on top of old content migrations nobody cleaned up. A chain becomes a loop when a URL eventually redirects back to itself, which triggers the browser “too many redirects” error and blocks the page entirely.

The fix is to collapse every chain to one hop. Point the first URL straight at the final destination with a single server-side 301 (or 308 for methods that must stay intact). Then update internal links, canonical tags, and the XML sitemap so they reference the final URL, not any intermediate one, so crawlers never enter the chain to begin with.