Canonical points to the homepage: fix a CMS deindexing bug

What it means

A page that is not the homepage has a rel=canonical pointing to the site root (e.g. https://example.com/). Google reads this as "treat the homepage as the authoritative version of this page" — so the page's own URL is not the one Google should index.

Why it matters

Google consolidates the interior page's link-equity and indexation signals into the homepage and treats the page URL as a non-canonical duplicate, dropping it from search results while the homepage gains nothing rankable for that content. At scale — when a theme or plugin generates the bug across many pages — a large share of a site's content silently disappears from the index. The usual cause is a CMS template emitting the canonical from a site-wide variable (get_site_url()) instead of the per-page permalink (get_permalink()), or an SEO plugin configured with a relative / canonical.

How to fix it

1. Export the full list of affected URLs from the report before changing anything.
2. Locate the canonical output in the template and fix it to compute from the current page's URL, not a global site-URL variable.
3. In WordPress, confirm the SEO plugin's canonical field is empty (self-referencing) or set to the page's own URL, and check header.php/functions.php for hardcoded canonical output.
4. Verify with curl -s URL | grep -i canonical on several pages, then request recrawl via Search Console URL Inspection.

When it's not a problem

There is no legitimate case for a non-homepage page to canonicalize to the homepage. If a page genuinely duplicates the homepage, redirect it or consolidate the content instead of using a canonical tag. The homepage's own self-canonical to root is correct and is not flagged.

How common is it?

1281 audited sites in our corpus currently show this issue. The breakdowns below show which platforms, gatekeepers, verticals and countries are most exposed.