A redirect loop is a redirect that eventually points back to itself. Unlike a redirect chain, which works but wastes resources, a loop resolves to nothing: the browser gives up, the crawler gives up, and the page is completely inaccessible.
That makes loops urgent rather than merely untidy. A looping URL is functionally deleted, and if it affects a template, an entire section disappears.
How they form
Conflicting rules in two places
The commonest cause. A redirect plugin sends A to B while a server rule sends B to A. Each was correct in isolation and nobody tested them together.
Protocol and www misconfiguration
A rule forcing HTTPS and another forcing non-www, applied in an order that never resolves. Classically produces a loop on the homepage, which is immediately obvious, or on a subset of paths, which is not.
Trailing slash rules fighting each other
One rule adds a slash, another removes it. The URL alternates indefinitely.
Case-sensitivity rules
A rule lowercasing URLs combined with a redirect map containing mixed-case entries.
Redirecting a URL to itself
Usually a data error in a bulk redirect import, where source and destination are identical. It looks harmless in a spreadsheet and breaks the URL entirely.
CDN and origin both redirecting
Edge rules and origin rules each handling the same path, which is easy to create and hard to see, because the origin behaves correctly when tested directly.
Finding them
- Browsers report them explicitly, usually as too many redirects.
- Search Console reports redirect errors in the Page indexing report.
- A crawl with redirect following will flag them, and this is the reliable way to find loops affecting only some paths.
- Test with and without www, with and without HTTPS, and with and without a trailing slash. Most loops appear in one of those combinations.
- Test through the CDN as well as directly against the origin, since the two can differ.
Fixing them
- Inventory every place redirects are configured: server, CDN, plugin, application. Loops almost always involve two of these.
- Consolidate to one mechanism where possible.
- Establish a single canonical form: protocol, www or not, trailing slash or not, and apply it in one rule.
- Check bulk imports for rows where source equals destination.
- Re-test the full matrix of variants after any change.
The systemic fix is having one place where redirects live. Loops are almost always a symptom of redirects being configured in two systems that do not know about each other.
Urgency
A loop on a single retired URL is minor. A loop on a template, a category path or the homepage removes those pages from search entirely and blocks visitors. If Search Console reports redirect errors at any volume, treat it as immediate rather than scheduling it.
Frequently asked questions
Why does the page load for me but not for others?
Caching. A cached response may serve locally while the loop occurs for anyone reaching the origin fresh. Test in a private window, and test through the CDN rather than only against the origin.
Do loops cause deindexing?
Effectively, yes. A URL that cannot be fetched cannot remain indexed indefinitely. Prolonged loops remove pages from search.
How quickly do pages recover after fixing?
As they are recrawled, typically days for important pages. The URLs return once they resolve to a real page again.
Can a loop affect only some pages?
Yes, and that is the harder case. Rules matching a path pattern can loop for that pattern while the rest of the site works, which is why a full crawl matters more than spot-checking the homepage.
Consulting CTA
If Search Console is reporting redirect errors or pages are inaccessible, book an SEO consultation.