What Is a Canonical URL?
Google defines the term in its documentation, What is canonicalization: “Canonicalization is the process of selecting the representative –canonical– URL of a piece of content. Consequently, a canonical URL is the URL of a page that Google chose as the most representative from a set of duplicate pages.” Google adds that this is “Often called deduplication”, and that it helps Google show only one version of duplicate content in its results.
So there are two canonicals to keep apart: the one you declare and the one Google selects. Google is explicit that your declaration is advice: “indicating a canonical preference is a hint, not a rule.” Duplicates are common and usually harmless. Google lists region variants, mobile and desktop versions, HTTP and HTTPS versions, sort and filter parameters, and demo sites accidentally left open to crawlers. It says some duplicate content “is normal and it's not a violation of Google's spam policies.”
Where rel="canonical" Comes From: RFC 6596
The rel="canonical" link relation is defined in RFC 6596, “The Canonical Link Relation”, published by the IETF in April 2012 by M. Ohye and J. Kupke. It is an Informational RFC, not a standards-track one. Its introduction says “‘canonical’ can be described informally as the author's preferred version of a resource.” Formally, it “specifies the preferred IRI from a set of resources that return the context IRI's content in duplicated form.” (An IRI is an internationalised URL.) The IANA registry entry reads: “Designates the preferred version of a resource (the IRI and its contents).”
The RFC sets one firm requirement: “The target (canonical) IRI MUST identify content that is either duplicative or a superset of the content at the context (referring) IRI.” It says search engines may then index content only from the canonical, consolidate properties such as link popularity onto it, and display it as the representative URL. Google says it supports rel canonical annotations “as described in RFC 6596” (Google).
A canonical link element in the <head> of a page looks like this:
<link rel="canonical" href="https://example.com/dresses/green-dresses" />For non-HTML files such as PDFs, the same signal can be sent as an HTTP header:
Link: <https://www.example.com/downloads/white-paper.pdf>; rel="canonical"How Google Chooses the Canonical URL
Google picks the canonical itself, using signals collected during indexing. Your rel="canonical" is one of those signals. According to Google's canonicalization page, when it indexes a page it works out the page's primary content. If several pages “seem to be the same or the primary content very similar”, Google clusters them. It then chooses the page that is “objectively the most complete and useful for search users” and marks it canonical.
Google names the factors involved: “whether the page is served over HTTP or HTTPS, redirects, presence of the URL in a sitemap, and rel="canonical" link annotations.” The choice has consequences:
- “The canonical page will be crawled most regularly; duplicates are crawled less frequently”.
- “Google uses the canonical page as the main source to evaluate content and quality.”
- A search result “usually points to the canonical page, unless one of the duplicates is explicitly better suited for a search user”, such as a mobile page for a mobile searcher.
The Canonical Signals You Control, Ranked by Google
Google ranks the three methods you control by strength, in How to specify a canonical URL: redirects and rel="canonical" are strong signals, and sitemap inclusion is a weak one. Google adds that the methods “can stack and thus become more effective when combined.” It also says none is required: “your site will likely do just fine without specifying a canonical preference.”
| Method | Google's description (verbatim) | Best used for | Limitation Google notes |
|---|---|---|---|
| Redirect (permanent) | “A strong signal that the target of the redirect should become canonical.” | Retiring a duplicate URL entirely | “Use this only when deprecating a duplicate page.” |
rel="canonical" link element or HTTP header | “A strong signal that the specified URL should become canonical.” | Duplicates that must stay reachable, such as tracking-parameter URLs | Element works only for HTML; use the HTTP header for PDFs. Use one method, not both. |
| Sitemap inclusion | “A weak signal that helps the URLs that are included in a sitemap become canonical.” | Large sites; listing only canonical URLs | “Less powerful signal to Google than the rel="canonical" mapping technique.” |
Classification from Google Search Central, How to specify a canonical URL (last updated July 10, 2026), not a measurement
Source: Google Search Central, How to specify a canonical URL with rel="canonical" and other methods, read October 10, 2026.
Google lists the three methods “in order of how strongly they can influence canonicalization”; bar lengths show that order only. Google says the methods can stack when combined.
Two other signals come from how the site is set up. Google “prefers HTTPS pages over equivalent HTTP pages as canonical” unless something conflicts, such as an invalid certificate or an HTTPS page that redirects to HTTP. It also prefers URLs that are part of hreflang clusters. Internal links count too: “When linking within your site, link to the canonical URL rather than a duplicate URL.”
Duplicate vs Near-Duplicate Pages
Canonicalization applies to exact duplicates and to pages whose main content is very similar. It does not apply to pages that are only about the same topic. Google's documentation refers to “duplicate or very similar pages” and clusters pages whose “primary content” is very similar. RFC 6596 allows the canonical to be a “superset” of the duplicate, such as a view-all page for a multi-page article, but not a different page.
Google's 2013 post 5 common mistakes with rel=canonical (Google marks it as possibly outdated) suggests a test. Put the two pages side by side and imagine you cannot read the language. Does “a very large percentage of the words” of the duplicate appear on the canonical? If the pages are “only topically similar”, it says, “the canonical designation might be disregarded by search engines.”
Language matters too. Google treats translated pages as duplicates only if the main content is in the same language. For regional variants in the same language, it says to use both canonicalization and hreflang. When Google wrongly clusters different pages together, its troubleshooting guide says the fix “boils down to ensuring that the pages that are clustered together are sufficiently different”. It warns that Google “might hold pages in a duplicate cluster for up to two weeks” after you fix the content.
Google-Selected vs User-Declared Canonical
Search Console shows both versions side by side in the URL Inspection tool. User-declared canonical is the canonical your page states, if it states one. Google-selected canonical is, in Google's words, “The page that Google selected as the canonical (authoritative) URL when it found similar pages on your site.” Google adds: “If the page has no alternate versions, the Google-selected canonical is the inspected URL.” The Google-selected value comes from indexed data only; a live test cannot predict it.
The Page indexing report groups URLs by the outcome. Three statuses involve canonicals:
| Status (as labelled by Google) | What Google says it means | Usually needs action? |
|---|---|---|
| Alternate page with proper canonical tag | The page is an alternate of another page and “correctly points to the canonical page, which is indexed” | No: “there is nothing you need to do” |
| Duplicate without user-selected canonical | A duplicate that names no canonical; Google chose another page and “will not serve this page in Search” | Only if Google picked the wrong page; Google calls it “working as intended” |
| Duplicate, Google chose different canonical than user | The page is marked canonical, “but Google thinks another URL makes a better canonical” | Check whether the pages really differ, and whether your signals conflict |
What Bing Says About Canonical Tags
Bing's Webmaster Guidelines say, under “Consolidate Duplicate URLs”, to “Avoid serving the same content across multiple URLs.” They warn that duplicate URLs “dilute signals and reduce Bing’s confidence in selecting a URL for grounding results or citations”. The recommended fix is to “Use canonical URLs, parameter controls, and consistent URL structures to consolidate signals”.
When content moves, the next section, “Handle URL Moves with Proper Redirects”, is direct: use 301 redirects for permanent changes, use 302s only for changes shorter than two days, and “Use redirects instead of canonical tags.” Bing's guidelines page does not describe how Bing weighs a canonical tag against other signals. We read it in a browser on October 7, 2026, because the page loads with JavaScript.
Common Canonical URL Mistakes
Most canonical problems come from conflicting signals or from a canonical that points at a page that is not a duplicate. The table pairs each common mistake with the source that describes it.
| Mistake | What goes wrong | Fix | Source |
|---|---|---|---|
| Canonical pointing to a different (non-duplicate) page, such as page 2 to page 1 or a category page to a featured article | The RFC says the target must be duplicative or a superset; content on the canonicalised page may be dropped | Use a self-referencing canonical on pages you want indexed | RFC 6596; Google, 2013 |
Mixed .html and extensionless URLs (/guide.html and /guide both return 200) | Two URLs serve one page; if the canonical, sitemap and internal links name different versions, the signals conflict | Pick one form and use it in the canonical, sitemap, internal links and structured data; redirect the other where possible | Google best practices |
| HTTP and HTTPS both live, or an HTTPS page with a canonical to HTTP | Google prefers HTTPS unless signals conflict; an HTTPS page canonicalised to HTTP is one such conflict | Redirect HTTP to HTTPS, point canonicals to HTTPS, and list HTTPS URLs in the sitemap | |
| Trailing slash and non-slash versions both return 200 | Google “treats each URL above separately”, so they are two URLs with the same content | Choose one, link to it consistently, and 301-redirect the other | Google, 2010 |
Different canonicals in different places (sitemap vs rel="canonical", or header vs element) | Google says not to specify different URLs as canonical for the same page using different techniques | Make every signal name the same URL | |
| Relative or malformed canonical URL | Google supports relative paths but does not recommend them; a URL missing its scheme is read as a path on your own site | Use absolute URLs, including https:// | Google; Google, 2013 |
Using robots.txt or noindex to pick a canonical | Google says not to use robots.txt for canonicalization and does not recommend noindex for it | Use rel="canonical" or redirects | |
| Canonical target that permanently redirects, errors or names another canonical | RFC 6596 says to avoid these targets; applications may ignore the declaration | Point to a live 200 page that is itself canonical | RFC 6596 |
The .html mismatch is easy to miss on static sites whose host serves both forms. On October 6, 2026, Media Morpheus changed the canonical, og:url and structured-data URLs of its guides from the .html form to the extensionless form its sitemap lists.
Canonical URL vs Related Terms
Canonical tags are often confused with the other ways of controlling which URL appears in search. Each does a different job.
| Tool | What it does | When to use it instead of a canonical |
|---|---|---|
| 301 redirect | Sends users and crawlers to another URL; a strong canonical signal | When the duplicate URL no longer needs to load for anyone |
noindex | Blocks a page from Search entirely | When the page should not appear at all; Google does not recommend it for choosing a canonical |
| robots.txt disallow | Controls crawling, not indexing | Not for canonicalization; Google may still index blocked URLs without their content |
hreflang | Marks language and regional alternates | Alongside a canonical, for same-language regional variants |
| Sitemap | Lists URLs you want crawled; a weak canonical signal | Always, listing canonical URLs only |
Common Misconceptions About Canonical URLs
| Myth | What the source actually says | Source |
|---|---|---|
| “A canonical tag forces Google to use my URL.” | “indicating a canonical preference is a hint, not a rule.” | |
| “Duplicate content gets a site penalised.” | Some duplicate content “is normal and it's not a violation of Google's spam policies.” | |
| “Every site must set canonicals.” | “none of them are required; your site will likely do just fine without specifying a canonical preference.” | |
| “A sitemap entry is as strong as a canonical tag.” | Sitemap inclusion is “A weak signal”; rel="canonical" is “A strong signal” | |
| “Canonical tags are the right way to handle a moved page.” | Bing: “Use redirects instead of canonical tags.” | Bing |
How to Check Your Canonical URL
Checking takes three looks: what your page declares, what Google selected, and whether your other signals agree.
- Read the declared canonical. View the page source and find the
<link rel="canonical">in the<head>; there should be exactly one. For the HTTP header version, runcurl -sI https://example.com/pageand look for aLink:line. Also request the.html, trailing-slash andhttp://variants of the URL and note which return 200 and which redirect. - Compare with Google's choice. Inspect the URL in Search Console and compare User-declared canonical with Google-selected canonical under Page indexing. Our guide on how to check if a page is indexed by Google walks through the tool.
- Check the Page indexing report for “Duplicate, Google chose different canonical than user”. If pages sit in “Discovered – currently not indexed” instead, that is a different problem; see how to fix “Discovered – currently not indexed”.
- Align the sitemap. List only the canonical form of each URL, then resubmit if it changed. Our guide on how to submit a sitemap to Google and Bing covers both consoles.
- Ask for re-evaluation of your most important fixed URLs with Request Indexing in URL Inspection. Google says the feature is subject to quotas.
Frequently Asked Questions
Is a canonical tag a directive or a hint?
A hint. Google says “indicating a canonical preference is a hint, not a rule”, and it may choose a different canonical for reasons such as content quality or technical signals.
Should a page have a self-referencing canonical?
Google recommends it: “Do include a rel="canonical" link on the canonical page itself (also known as a self-referential canonical).” RFC 6596 also allows the target to be self-referential.
Can a canonical point to another domain?
Yes. RFC 6596 says the canonical may exist on a different hostname or domain. Google's troubleshooting guide adds that if the canonical is in a Search Console property you don't own, you won't see the duplicate page's traffic.
What happens if a page has two canonical tags?
Google's 2013 guidance says that when more than one rel=canonical is specified, Google will likely ignore all of them. RFC 6596 also says to specify only one canonical link relation per resource.
Should I use a canonical tag for syndicated articles?
Google does not recommend it for avoiding duplication by syndication partners, because the pages are often very different. Its troubleshooting guide says the most effective option is for partners to block indexing of your content.
Can I set a canonical for a PDF?
Yes, with a rel="canonical" HTTP Link header, since the link element works only in HTML. Google says it supports the header method for web search results only.
How long does Google take to apply a canonical change?
Google gives no fixed time. Its troubleshooting guide says it might hold pages in a duplicate cluster for up to two weeks after content fixes, and suggests Request Indexing for the most important URLs.
Does Google support canonical tags added with JavaScript?
Yes, but Google prefers the canonical in the HTML source and asks that JavaScript not change it. If you cannot set it in HTML, leave it out of the source and set it only with JavaScript.
Does Bing use canonical tags?
Bing's Webmaster Guidelines tell site owners to use canonical URLs to consolidate duplicate URLs, and to use 301 redirects rather than canonical tags when content moves permanently.
Sources and Methodology
This explainer is built from Google Search Central and Search Console documentation, RFC 6596 and Bing's Webmaster Guidelines, all read on October 7, 2026. Bing's page was read in a browser because it loads with JavaScript. Two older Google blog posts (2010 and 2013) are cited for examples, and Google itself flags the 2013 post as possibly outdated. Definitions and signal-strength descriptions are quoted verbatim. We did not run canonicalization tests for this explainer. The one first-party fact, the Media Morpheus URL fix on October 6, 2026, is stated without any claim about its effect.
- official Google Search Central, What is canonicalization. Read October 7, 2026.
- official Google Search Central, How to specify a canonical URL with rel="canonical" and other methods. Read October 7, 2026.
- official Google Search Central, Fix canonicalization issues. Read October 7, 2026.
- official Google Search Console Help, URL Inspection tool. Read October 7, 2026.
- official Google Search Console Help, Page indexing report. Read October 7, 2026.
- official M. Ohye and J. Kupke, IETF, RFC 6596: The Canonical Link Relation, April 2012. Read October 7, 2026.
- official Bing Webmaster Tools Help, Webmaster Guidelines, sections 6 and 7. Read October 7, 2026.
- official Maile Ohye, Google Search Central Blog, To slash or not to slash, April 21, 2010. Read October 7, 2026.
- official Allan Scott, Google Search Central Blog, 5 common mistakes with rel=canonical, April 8, 2013. Read October 7, 2026.