Quick answer: Canonical tags are HTML link elements that identify the preferred URL for duplicate or substantially similar pages. They help search engines consolidate signals, such as inbound links, around a representative URL. A declared canonical is a strong hint, not a directive: Google may select a different URL, and users are not redirected.
Key facts
- Google can choose a different canonical from the URL you declare. What is canonicalization
- Canonicalization can consolidate inbound-link signals from duplicate URLs. How to specify a canonical URL with rel="canonical" and other methods
- Redirects and canonical annotations are strong signals; sitemap inclusion is weaker. How to specify a canonical URL with rel="canonical" and other methods
- HTML canonical elements belong in the document head; absolute URLs are recommended. How to specify a canonical URL with rel="canonical" and other methods
- Search Console URL Inspection reports Google’s selected canonical. Fix canonicalization issues
TL;DR: Point genuinely equivalent pages directly to a preferred, accessible canonical URL, keep supporting signals consistent, and verify Google’s selection in Search Console.
- Canonical URL
- The representative URL selected from a group of duplicate or substantially similar pages.
- Canonical tag
- An HTML link element using rel="canonical" to declare the preferred URL for a page’s content.
- Self-referencing canonical
- A canonical annotation that points to the preferred URL of the page on which it appears.
- Google-selected canonical
- The URL Google chooses as representative, which can differ from the site’s declared canonical.

What are canonical tags?
Canonical tags are HTML link elements that name a preferred URL. Use them when several URLs show the same or very similar content. A tag is a hint, not a command. Google may choose a different URL, and the tag does not ensure indexing (Google Search Central). It does not redirect visitors.
The tag, URL and selection process
- Canonical tag: an HTML link element with rel=”canonical” and a preferred URL.
- Canonical URL: the address chosen to represent a set of matching pages. The user-declared canonical is your choice. The Google-selected canonical is Google’s choice.
- Self-referencing canonical: a canonical annotation whose target is the URL of the page that contains it.
- Canonicalization: the process of choosing a URL to represent matching pages.
This choice is part of how search engines crawl, index and rank pages. It does not control a visitor’s browser.
A product URL example
A campaign URL may show the same product as a clean URL. If the content is the same, the campaign page can name the clean address as its preferred URL.
Campaign URL:
https://example.com/product?utm_campaign=spring
HTML tag in that page’s head:
<link rel=”canonical” href=”https://example.com/product”>
The tag names the clean product URL. Visitors still see the campaign address in their browser.
Why canonical tags matter for SEO
The main benefit is to bring SEO signals together. When duplicate URLs receive inbound links, canonicalization helps Google combine those signals around the chosen URL (Google Search Central).
For example, both a clean product URL and a tracking URL may receive links. Choosing a canonical helps Google treat those links as signals for the same page.
This is not an automatic ranking boost. It is also not a way to combine the value of unrelated pages. A distinct article and a product page do not become duplicates just because one names the other as its canonical.
Ordinary duplicate content does not break Google’s spam rules. Canonical tags are not a way to avoid a general duplicate-content penalty (Google Search Central).
Canonical tags do not block crawling. Do not treat them as a direct control over crawlability and crawl budget. Use them to make your preferred URL clear, as part of a wider technical SEO plan.
Which URLs should share a canonical?
Pages should share a canonical when their main content is the same or very similar. A shared topic is not enough. Compare the pages before you choose a preferred URL (Google Search Central).
- Compare the main content. Check the product, text and results on each page. Tracking parameters may change how a visit is recorded without changing the content.
- Review meaningful changes. Color, size and filter parameters need a closer look. They may show distinct products or useful sets of results. Do not strip all parameters by default.
- Choose a stable target. Pick a URL that works, can be indexed and shows the intended content. Keep your preferred host and protocol consistent. Do not choose a page just because it has more links.
A campaign URL can share the clean URL’s canonical when only the tracking details differ, as in the product example above.
A canonical tag lets matching URLs stay available. A permanent redirect instead sends visitors from an old URL to its replacement.
Redirects and rel=”canonical” are both strong signals. Keep them aligned (Google Search Central).
How do you implement canonical tags?
Add the preferred URL to the page’s HTML head. For a PDF, use an HTTP response header. Point duplicate pages to the preferred page, and give that page a self-referencing canonical. Then check what your site sends, not just its settings.
Add the HTML link element
Google accepts HTML canonical tags only in the document’s head. It recommends absolute URLs and a self-referencing canonical on the preferred page (Google Search Central).
An absolute URL includes the protocol and host. Use your site’s preferred version. The examples below use HTTPS.
Duplicate page URL:
https://example.com/guide?print=yes
HTML to place inside its head:
<link rel=”canonical” href=”https://example.com/guide”>
Preferred page URL:
https://example.com/guide
Self-referencing HTML to place inside its head:
<link rel=”canonical” href=”https://example.com/guide”>
The tags match. On the preferred page, the target is that page’s own URL. This is a recommended practice, not a requirement for indexing.
Use a response header for PDFs
For non-HTML files such as PDFs, Google supports a canonical URL in an HTTP Link response header (Google Search Central).
HTTP response header:
Link: <https://example.com/guide.pdf>; rel=”canonical”
Set this in the server’s response headers, not in the page text.
Check the delivered declaration
Keep one consistent canonical target per document. Check for conflicts between templates, plugins and response headers. If you use WordPress and Yoast SEO, inspect the output rather than trusting the plugin settings.
Where practical, put the tag in the source HTML. For tags added or changed by scripts, see JavaScript SEO and Rendering.
Keep canonical signals consistent
Use the same preferred URL across your tags, redirects and sitemap. Signals that agree can reinforce each other. Google treats redirects and canonical tags as strong signals, while sitemap entries are weaker (Google’s guidance).
Align declarations with the final URL
Point each duplicate to the final preferred URL. That page should also point to itself.
| Page URL | Canonical target |
|---|---|
| https://example.com/item?view=print | https://example.com/item |
| https://example.com/item?ref=email | https://example.com/item |
| https://example.com/item | https://example.com/item |
- Use a working target. Avoid URLs that are unavailable, redirect elsewhere or have a noindex rule.
- Align existing redirects. They should support the same preferred URL. You do not need to add redirects everywhere.
- Check your sitemap. List preferred URLs in XML sitemaps.
- Allow access to the tag. Robots.txt blocking can stop Google from fetching it.
- Keep noindex separate. A canonical names a preferred URL, while noindex controls index exclusion. Do not use noindex to force a canonical choice.
Common canonical mistakes to avoid
Watch for tags that send mixed signals or group pages that should stay separate.
- Loops: Do not make page A point to B while B points to A.
- Chains: Point each duplicate straight to the preferred URL, not through other URLs.
- Conflicting targets: Check that templates, plugins and response headers name the same target for a page.
- Tags in the wrong place: Keep HTML canonical tags in the head. Google accepts them only there (Google Search Central).
- Homepage defaults: Do not let a shared template point distinct products, articles or category pages to the homepage.
Each page in a paginated sequence should have its own canonical URL. Do not point them all to the opening page. Google no longer uses rel=”prev” or rel=”next” to link pages in these sequences (Google Search Central).
Do not group all translations under one language’s canonical. See International SEO and Hreflang for guidance on language versions.
For syndicated copies, Google does not recommend canonical tags to prevent duplication. It identifies partners blocking indexing of their copies as the most effective solution (Google Search Central).
How do you check whether Google accepted your canonical?
Use Search Console’s URL Inspection tool to check Google’s choice. Compare the user-declared canonical with the Google-selected canonical in the indexed page data (Google Search Central).
Finding a tag only proves that the tag exists. Neither that check nor the live test proves that Google chose its target.
- Check the setup. Look for rel=”canonical” in the HTML head. Check response headers too, including those for PDFs. A crawl with Screaming Frog SEO Spider can help find conflicts, chains and broken targets. Missing tags need review; they are not always errors. Include these checks in your technical SEO audit.
- Check Google’s choice. In Google Search Console, inspect a sample of duplicate URLs and their intended targets. Compare the canonical fields in the indexed data.
- Find the cause of differences. Compare the pages’ main content. Check that the target works and that all tags agree. Review internal linking to see which URLs your site favors.
- Check again after changes. Google must revisit and process the pages. Track fixed errors and whether Google’s choice matches yours. Do not use tag counts as proof of ranking gains.
Frequently asked questions
Are canonical tags necessary on every page?
No. They are useful when several URLs show the same content. Google recommends a self-referencing canonical on the preferred page, but a tag is not required for every page to be indexed (Google Search Central).
How long does Google take to process a changed canonical tag?
Do not expect an immediate switch or rely on a fixed turnaround. Google needs to revisit the pages and assess its signals. Use URL Inspection to check which URL it selected (Google Search Central).
Should syndicated copies on another domain use canonical tags?
For this case, follow the partner-side indexing guidance in “Common canonical mistakes to avoid.”
Do canonical tags affect campaign tracking in analytics?
A canonical tag does not redirect visitors or remove tracking parameters from their browser’s address. A campaign URL can name a clean canonical URL and still keep its parameters. How analytics records the visit depends on your tracking setup and any redirects, not on the tag alone.