How to Check Which Canonical Google Chose
You can declare a canonical URL, but Google makes the final choice. When two similar pages exist, the canonical Google selects decides which one appears in search. This guide shows where to read both values, how to interpret a mismatch, and what to change when Google disagrees with you.
- Read both canonicals
- Why they differ
- What the live test can’t tell you
- Fix steps
User-declared canonical: /guide/ Google-selected canonical: /guide/ (match) mismatch? inspect both pages
Inspect the URL in Search Console and open Page indexing. Compare User-declared canonical with Google-selected canonical. Google says the first is only considered, not guaranteed to be honoured, and that canonical selection happens at indexing time, so the live test cannot predict it. Check the indexed result, not the live test.
The two canonical values
Google’s URL Inspection help defines them. The user-declared canonical is the canonical URL the page declares through a rel="canonical" link, an HTTP header, a sitemap or another method; “None” is fine for non-alternate pages, and Google considers it but does not guarantee to honour it. The Google-selected canonical is the URL Google chose as authoritative when it found similar pages. If the page has no alternates, it equals the inspected URL.
Where to look
- Open Search Console and paste the full URL into the search bar at the top.
- Read the indexed view (not the live test).
- Expand Page indexing and find “User-declared canonical” and “Google-selected canonical”.
Google notes the Google-selected value may be a few hours out of date. See also the URL Inspection tool guide.
Reading the result
| You see | Meaning | Action |
|---|---|---|
| Both equal the inspected URL | Google keeps this page as canonical | None |
| Both point to another URL | This page is a duplicate; the other is preferred | Fine if intended |
| Declared is this URL, selected is another | Google disagrees with you | Compare the two pages |
| Declared is “None”, selected is another | Google chose without your input | Declare a canonical if you disagree |
The last two match the status names in the Page indexing report: Duplicate without user-selected canonical and Duplicate, Google chose different canonical than user.
Why Google may choose differently
Google’s Page indexing help says that if your declared canonical is not similar to the page, Google will not choose it, so you should correct the canonical or make the content meaningfully distinct. Google’s duplicate URL documentation also lists signals it uses: redirects and rel="canonical" are strong signals, sitemap inclusion is weak, and Google generally prefers HTTPS over HTTP.
Signals that help your preferred URL win
| Signal | Strength per Google | Practice |
|---|---|---|
| Redirect | Strong | Use only when deprecating the duplicate |
| rel=”canonical” | Strong | Self-referencing on the canonical |
| Sitemap inclusion | Weak | List canonical URLs only |
| Internal links | Recommended | Link to the canonical URL |
| HTTPS | Preferred over HTTP | Redirect http to https |
Google also advises absolute URLs for canonicals, no fragments, and not using robots.txt or the removal tool for canonicalisation. See canonical tags and duplicate content.
Step by step: diagnose a mismatch
Inspect the problem URL
Paste it into the top bar. Use the indexed view.
Write down both canonicals
Copy the declared and selected URLs.
Inspect the selected URL
Confirm it is indexed and returns 200.
Compare the two pages
Check content, title, headings and links. Are they near duplicates?
Decide which URL should win
If Google’s choice is fine, update your canonical and links to match. If not, make your page distinct or consolidate.
Align signals and wait
Update canonical, sitemap and internal links. Recheck after recrawl.
Quick extra check: the site: operator (my own tip)
Searching site:yourdomain.com/page-slug shows which version appears in results. It is a rough check only, because operator results are not a complete index listing. See Google search operators and treat URL Inspection as the authority.
Worked example (illustrative)
| Inspected URL | Declared / Selected | Fix |
|---|---|---|
| /seo-guide/ | /seo-guide/ / /seo-guide/ | None |
| /seo-guide/?ref=news | /seo-guide/ / /seo-guide/ | None (parameter alternate) |
| /seo-tips/ | /seo-tips/ / /seo-guide/ | Differentiate or merge into /seo-guide/ |
| /pricing-2/ | None / /pricing/ | Redirect or declare canonical |
Notes for WordPress sites
- SEO plugins add self-referencing canonicals. Check the page source for exactly one canonical tag.
- Duplicate pages often come from copies like “-2” slugs. Merge or redirect them (duplicate content in WordPress).
- Pagination, tags and category archives can look similar; set their canonicals and indexing deliberately.
- Clear page cache so Google reads the updated tag.
Troubleshooting
| Symptom | Likely cause (my own diagnosis) | Action |
|---|---|---|
| Selected differs from declared | Pages too similar or conflicting signals | Differentiate or align signals |
| Two canonical tags in source | Theme plus plugin | Remove one source |
| Live test looks fine, indexed differs | Selection happens at index time | Wait for recrawl |
| Canonical to a redirecting URL | Outdated canonical | Point to the final URL |
Checklist
Check
- Inspect indexed view
- Note both canonicals
- Inspect the target
- Compare content
Signals
- Self-referencing canonical
- Absolute URLs
- Canonical in sitemap
- Internal links to canonical
Fix
- Merge duplicates
- Redirect old copies
- Make pages distinct
- Clear cache
Follow up
- Request indexing
- Recheck in days
- Log changes
- Audit quarterly
What Google’s pages do not say
Google explains the two canonical fields, that selection is made at indexing time, and the relative strength of signals. It does not give a formula for how similar pages must be, or a time for canonical changes to appear. The extra site: check, the example and the troubleshooting are my own practice.
Key terms
- Canonical
- Preferred URL among duplicates.
- User-declared
- The canonical you specified.
- Google-selected
- The canonical Google chose.
- Duplicate
- Near-identical content at two URLs.
Common mistakes
Trusting the live test for canonicals
It cannot predict selection.
Canonicalising unlike pages
Google ignores it.
Blocking duplicates in robots.txt
Not a canonical method.
Noindex to choose a canonical
Google advises canonical.
Relative canonical URLs
Easy to break.
Sitemap lists duplicates
Mixed signals.
Related guides
Google-selected canonical questions
Does Google always honour my canonical?
No. Google considers it but does not guarantee to honour it.
Can the live test show the selected canonical?
No. Selection happens at indexing time, so the live test cannot predict it.
What if the two values differ?
Compare the pages. Make them distinct, consolidate, or align your signals with Google’s choice.
Is a mismatch an error?
Not necessarily. It matters only if the wrong page is preferred.
Should I put canonicals in my sitemap?
Yes, list canonical URLs. Google calls sitemap inclusion a weak signal.
Do I need a self-referencing canonical?
Google recommends one on the canonical page.
Can I use robots.txt to canonicalise?
No. Google advises against it.
How long until a change shows?
Google gives no fixed time; recheck after recrawl.
- Google Search Console Help: URL Inspection tool
- Google Search Central: Consolidate duplicate URLs
- Google Search Console Help: Page indexing report
Field definitions, signal strengths and canonical rules follow Google’s pages. The site: tip, worked example, troubleshooting and WordPress notes are my own practice.




