How to fix canonical tag problems: common mistakes and how to find them
A canonical tag tells search engines which URL you prefer. The mistakes that make them ignore it or pick the wrong page, and how to find them on your site.
The same content often lives at several URLs: with and without www, with tracking parameters, sorted or filtered, printed, or syndicated. A canonical tag — <link rel="canonical" href="https://example.com/shoes/"> in the page's <head> — tells search engines which of those URLs you'd like them to show. Search engines treat it as a strong hint, not an order, and they ignore it when it contradicts other signals.
Most canonical problems are template mistakes that repeat across hundreds of pages. Here's how to recognise and fix them.
What a good canonical looks like
- One canonical tag per page, in the
<head>. - An absolute URL with the right protocol and host:
https://example.com/shoes/, not/shoes/. - On a normal page, it points to itself (a self-referencing canonical).
- It points to a URL that returns 200, is indexable and doesn't redirect.
- It agrees with your other signals: internal links, the sitemap and redirects all use the same preferred URL.
Common canonical problems
The canonical points to a redirect, a 404 or an error
After a URL change, templates often keep emitting the old address. Now every page says "prefer this other URL", and that URL redirects or no longer exists. Search engines usually fall back to choosing a canonical themselves, and you lose control of which URL appears. Update the canonical to the final 200 URL.
noindex combined with a canonical to another page
noindex says "don't show this page"; a canonical to another URL says "this page is a duplicate of that one". Together they send mixed signals. Decide which you mean: if the page is a duplicate, keep the canonical and drop noindex; if it should simply stay out of search, use noindex with a self-referencing canonical or none. Our guide on what noindex means covers the difference.
The canonical points to a noindex page
Telling search engines to prefer a URL that itself says "don't index me" leaves them with nothing to show. Point the canonical to an indexable page, or remove noindex from the target.
Every page canonicalises to the homepage
A theme or plugin setting that outputs the site's root URL as the canonical on every page effectively asks search engines to drop all your other pages. It's rarer than it used to be, but it still happens after migrations. Check the source of a few inner pages.
Relative, malformed or multiple canonicals
A relative href is technically allowed but fragile. Two canonical tags that disagree — often one from the theme and one from an SEO plugin — are typically both ignored. Keep exactly one, generated in one place.
Protocol, host and trailing-slash mismatches
If your site runs on https://www.example.com/ but canonicals say http://example.com/page, each points to a URL that redirects. Pick one host, one protocol and one trailing-slash convention, and use it in canonicals, internal links and the sitemap.
Cross-domain canonicals
A canonical to another domain is legitimate when content is deliberately syndicated — a partner republishing your article can point to your original. On your own site it's usually a mistake, for example a staging or old domain left in a template.
Paginated series pointing to page one
Page 2 of a category lists different items from page 1, so it isn't a duplicate. Give each page in a series its own self-referencing canonical, unless you offer a genuine "view all" page.
Check a page's canonical
Rudra's free SEO checker reads the canonical tag, checks that it's a single absolute URL and, when it points elsewhere, requests the target to see whether it answers 200.
How to find canonical problems
On one page
View the page source and search for rel="canonical", or run the page through Rudra's free SEO checker. It reports a missing, relative or duplicated canonical and a canonical that points away from the page. When the canonical differs from the page's final URL, it sends one HEAD request to the target to check its status, and it flags a noindex page whose canonical points elsewhere as a conflict. A canonical to another domain is noted for information, because it can be intentional.
Across the site
Canonical problems are really relationships between pages, so a crawl finds them best. A Rudra site scan (create a free account and start one from your dashboard; the free website audit covers a single page) compares each page's canonical with what the crawl saw at that address and groups problems by reason: the canonical target redirects, returns an error status, is noindex or sits on another host, or a noindex page canonicalises to another URL. Targets the crawl didn't reach, because of page or depth limits, are counted as unverified rather than reported as broken.
In Google Search Console
The Page indexing report lists pages marked "Duplicate, Google chose different canonical than user" and "Alternate page with proper canonical tag". URL Inspection shows both the canonical you declared and the one Google selected. When they differ, look for conflicting signals: internal links, redirects or the sitemap pointing to the other URL.
Fixing them in the right order
- Choose the preferred form of every URL (protocol, host, trailing slash, parameters).
- Redirect the other forms to it with 301s where possible — a redirect is a stronger signal than a canonical.
- Generate one absolute, self-referencing canonical per page from a single template or plugin.
- Update internal links and the sitemap to use the preferred URLs only. See how to diagnose sitemap problems.
- For genuine duplicates you must keep (filters, print versions), point their canonical to the main version.
- Re-crawl, then watch Search Console over the following weeks as pages are recrawled.
Frequently asked questions
Does every page need a canonical tag?
It's not required, but a self-referencing canonical on every indexable page is good practice. It protects you against parameter and tracking-code variants you don't control.
Will Google always follow my canonical tag?
No. It's a strong hint that Google weighs against redirects, internal links, sitemaps and content similarity. When signals conflict, Google may choose a different canonical, which URL Inspection in Search Console will show.
Should I use a canonical or a 301 redirect?
Use a 301 when visitors don't need the duplicate URL at all, such as an old address or the non-www version. Use a canonical when the duplicate must stay accessible, such as a filtered or sorted listing.
Can a canonical point to a different domain?
Yes, for deliberate syndication, where another site republishes your content and credits the original. On your own site, a canonical to another domain is usually a leftover from staging or a migration.