A link leads to a page that names another as the main one
In the reportInternal links point to a canonicalized page
internal_link_to_canonicalized_page · Review item
This appears when links on your pages lead to a page whose canonical names a different page, one that works and names itself. The link sends visitors and crawlers to a version you told search engines is not the main one. Link to the main page instead.
Severity: Low. Review it when convenient; it is a consistency question, not an error.
Why it matters
Google advises that when linking within your site you link to the canonical URL rather than a duplicate, because linking consistently to the canonical helps it understand your preference.
What to do
- Look at the finding: it shows the linked page, its canonical and up to ten links to it.
- Change those links to the canonical address.
- If the linked page should be the main one, change its canonical to point to itself.
When to leave it as it is: if the linked version is intended for navigation, filtering or tracking, and its canonical is right, you can leave the links.
What we read
We read the <a href> links to your own host in the HTML of each crawled page, as for the other link checks, and follow each target's redirects. A target is reported when it answered 200 with HTML, carries no noindex, is not listed in your sitemap, and declares exactly one canonical on your own site that differs from it.
The canonical page must itself have been fetched and answer 200 directly, without noindex or a block, and name itself as canonical. Other cases are left to their own checks, and a canonical that only drops tracking parameters, such as utm_source, is not reported.
Example & evidence
Illustrative example — one link, the page it leads to and that page's canonical, not a live audit.
The homepage https://bakery.example/ contains:
<main><a href="/bread/sourdough-loaf/">Sourdough</a></main>/bread/sourdough-loaf/ answers 200 with:
<link rel="canonical" href="https://bakery.example/bread/sourdough/">and /bread/sourdough/ answers 200 and names itself, so the linked page is reported with its canonical and the link.
A possible correction: change the link to /bread/sourdough/.
Confirm & check again
Check for yourself: open the linked page's source and read its rel="canonical", check its response headers for a canonical Link header, and compare the address with the one in your link.
Check again: Full audit — run a new audit of the site. The old finding for this target closes only when it is not reported again, the check gives a usable result, and every page that linked to the target has complete evidence: it is read again as complete HTML that does not rely on scripts, with conclusive results for the current links on it that this check covers, and no longer links to this page canonicalized away, or it now answers directly with 200 and a file that is not HTML, or starts with a permanent redirect whose whole chain stays within your site. It may also close when another check takes over the same target, which does not by itself show the problem is fixed. Links pointed to another page canonicalized away may produce a new finding, if that page meets this check's conditions.
Limits & sources
The check does not show which address Google chose, and canonicals added only by JavaScript, and canonical links in the HTML outside the page's head, are not seen; a canonical in the Link header is read. Canonical pages beyond the 20 fetched separately per audit are not checked.
Confidence: High because the link, the canonical and the canonical page are all read directly.
Sources
- Google, How to specify a canonical URL with rel="canonical" and other methods — advises linking to the canonical URL rather than a duplicate when linking within your site. Checked 25 September 2026.
Editorial review: 29 September 2026