en
/login
Koval SEO

A canonical target redirects

In the reportCanonical target redirects

canonical_target_redirect · Review item

Short answer

Pages name an address on your site as canonical, but that requested address redirects before reaching a final HTTP 200 response. Check the recorded response and final URL, then point the canonicals directly to the intended main page.

Severity: Medium. Review the redirect and intended destination. Both permanent and temporary redirects are review items; a temporary redirect can be intentional.

Why it matters

Google treats both a canonical and a redirect as strong signals, and advises against naming different URLs for the same page through different methods. The canonical link standard asks not to name the source of a permanent redirect as a canonical; it allows the source of a temporary one.

What to do

  1. Read the requested target, final URL, redirect-status evidence and source pages in the finding.
  2. Check that the final page is the intended main version and is suitable for indexing before changing the source pages’ canonicals.
  3. If the redirect is temporary and the target will return, leave it only after reviewing that intention.

When to leave it as it is: A temporary redirect during maintenance can be intentional if the named address will answer directly again. Review the final page rather than assuming its HTTP 200 response proves suitability.

What we read

We group source pages by their single other same-site canonical target. We reuse retained crawl observations and the existing bounded canonical-target probes, with the run’s frozen request and probe limits. No new acquisition is introduced by this rule.

The requested target must have a settled final HTTP 200 response without a confirmed error and either recorded redirect hops or a different final URL. The first hop distinguishes permanent redirects (301 or 308) from temporary or unclassified ones. A final noindex or onward canonical does not suppress this redirect finding.

Example & evidence

Illustrative example — one page and the address it names, not a live audit.

https://bakery.example/cakes/?sort=price declares:

<link rel="canonical" href="https://bakery.example/cakes-old/">
GET https://bakery.example/cakes-old/
HTTP/1.1 301 Moved Permanently
Location: https://bakery.example/cakes/

GET https://bakery.example/cakes/
HTTP/1.1 200 OK

The requested target redirects before the final page answers, so the finding records its redirect status and final URL.

A possible correction: change the canonical to https://bakery.example/cakes/, which answers 200 directly.

Confirm & check again

Check for yourself: Request the exact address named in the canonical with a redirect checker. Compare its first redirect status and final URL with the recorded evidence.

Check again: Full audit — Run a new site audit. A sufficient clean recheck of the same requested target can close the finding. If source pages stop naming it, every previous source needs a sufficient recheck to prove removal; the old target then need not be fetched. An unread target or unread or partial source pages cannot prove a fix. Changing to a different redirecting target creates a separate finding, while dismissal stays independent of technical state.

Limits & sources

Only observed HTTP redirects count. Script or meta-refresh navigation is not followed. Unobserved, denied, budget-limited, rate-limited or unsettled targets remain unknown; confirmed errors belong to the target-error check. A page reached as the destination of another address’s redirect is not itself treated as redirecting. A final HTTP 200 response does not establish indexability or Google’s canonical choice.

Confidence: High The requested target’s redirect and final response are observed. The first hop’s status changes the explanation, not whether the redirect is reported.

Sources

Editorial review: 2 November 2026