What it checks
We request the page you gave us, follow up to five redirects, and record the final status code and every hop on the way. The check runs from outside your network, with the same passive request a crawler would make.
Why it matters for search engines
Search engines index the page a URL finally resolves to. A 404 or 410 drops the page from the index; repeated 5xx responses slow crawling and, if they persist, remove the page. A long redirect chain wastes crawl budget and can lose the target entirely if one hop breaks.
Why it matters for AI assistants
AI crawlers fetch a URL once and take what they get. A 5xx or a redirect loop at that moment means the page is simply absent from that assistant's answer. There is no retry queue you can rely on.
How to fix it
Aim for one hop at most: the plain http host to https, or www to apex, then 200. Fix the origin for 5xx responses, and give permanently moved pages a single 301 to the new address rather than a chain.
HTTP 200 after 302 redirect to https://www.theguardian.com/international.What quietwatch reports
ok for 2xx after redirects. fail for 4xx, 5xx, no response, or more than five redirects, with the last hop named. unknown when robots.txt disallowed our bot from fetching the page at all.
Sources
- Google, HTTP status codes and network errors in Search
- RFC 9110, HTTP Semantics, section 15 (status codes)
All twelve checks · Previous: AI crawlers · Next: Time to first byte