---
metadata:
- name: generator
content: Diplodoc Platform v5.55.0
alternate:
- https://yandex.com/support/webmaster/en/robot-workings/canonical.md
- https://yandex.com/support/webmaster/tr/robot-workings/canonical.md
- href: en/robot-workings/canonical.md
type: text/markdown
title: Markdown version
- href: ../llms.txt
type: text/markdown
title: llms.txt
---
> **Documentation Index:** Fetch the complete configuration index at https://yandex.com/support/webmaster/en/llms.txt
# Canonical URLs
If a site has a page available at multiple URLs, or pages with identical or similar content, the Yandex robot may count them as [duplicates](https://yandex.com/support/webmaster/en/robot-workings/double.md). In this case, it will combine the pages in a group of duplicates and choose one of them, the most informative and relevant to the search query, to be displayed in the search results. Such page is called _canonical_.
You can use the [rel="canonical" attribute](#set-canonical) to specify which page to show in the search results. You can also specify the canonical URL to change the [site address](#move) to a domain with or without the `www` prefix.
If you want to change the main site address when moving from HTTP to HTTPS, use the [301/302 redirect](https://yandex.com/support/webmaster/en/robot-workings/managing-redirects.md). The attribute `rel="canonical"` for addresses of this format is no longer supported. For more information, see the article [Disabling rel="canonical" support when moving from HTTPS to HTTP](https://webmaster.yandex.ru/blog/otklyuchenie-podderzhki-rel-canonical-pri-pereezde-s-https-na-http) in the Yandex blog for webmasters.
{% note alert %}
The Yandex robot interprets references to the canonical address as recommendations and can ignore them [in several cases](#not-use).
{% endnote %}
## How do I specify the canonical URL of a page? {#set-canonical}
Add the canonical URL in the `rel="canonical"` attribute using one of the following methods:
{% list tabs %}
- HTML code
For example, the page is accessible at two addresses: `www.example.com/pages?id==2` and `www.example.com/blog`.
If the preferred address is `/blog`, add the `link` element to the HTML code of the `/pages?id=2` page:
```html
```
- HTTP header
Let's say a site has a PDF file available at multiple URLs: `www.example.com/offer/file.pdf` and `www.example.com/files/file.pdf`. If the preferred URL is `/offer/file.pdf`, configure the server to pass the following in the HTTP header of the `/files/file.pdf` page:
```http
Link: ; rel="canonical"
```
{% endlist %}
{% note info %}
Specify the canonical URL within the same domain. Set the canonical URL as the absolute path, for example `http://example.com/blog/`.
{% endnote %}
The page with the `rel="canonical"` attribute that points to another URL is considered _non-canonical_.
The robot learns about the changes when it crawls the site. If the canonical URL is entered correctly and [the robot doesn't ignore the instructions](#content), the non-canonical page disappears from the search results. To see if the page was removed from search results, go to **Indexing** → **Searchable pages** (**Excluded pages** section) in Yandex Webmaster.
{#content}
The robot ignores instructions if the contents of the canonical and non-canonical page significantly differ. In this case, a non-canonical page may be included in the search. To check this, go to **Indexing** → **Searchable pages**.
To exclude a non-canonical page that contains GET parameters or tags (UTM, from, and so on) in the URL, add the [Clean-param](https://yandex.com/support/webmaster/en/robot-workings/clean-param.md) directive in the [robots.txt](https://yandex.com/support/webmaster/en/controlling-robot/robots-txt.md) file. Otherwise, use the [Disallow](https://yandex.com/support/webmaster/en/robot-workings/allow-disallow.md#disallow) directive.
## How to change the site address URL using the canonical URL {#move}
You can specify the canonical URL to change the site address to a domain with or without the `www` prefix.
The bot will interpret the canonical address as a redirect to the new [main site address](https://yandex.com/support/webmaster/en/yandex-indexing/site-addresses.md) and group the two site versions. To do this, add a link to the new site page with the `rel="canonical"` attribute in the HTML code or in the HTTP header of all old site pages. For example, you want to change `https://www.example.com` to `https://example.com`. On the page `https://example.com/main/`, you'll need to specify:
```html
```
If the attribute points to a different page, the robot might consider this a difference in the site structure. In this case, the site can't be moved. If the attribute is only added to some pages, it will not point to the main site address.
If you change the URL, make sure that the contents match on the old site and new site. For more information, see [relocation instructions](https://yandex.com/support/webmaster/en/yandex-indexing/moving-site.md).
{% note tip %}
To improve the chances that both users and indexing robots will land on the preferred version of your site, use the [301/302 redirect](https://yandex.com/support/webmaster/en/robot-workings/managing-redirects.md).
{% endnote %}
## Cases where the canonical address isn't taken into account {#not-use}
The Yandex robot doesn't consider the URL canonical if:
- At the time of crawling, non-canonical pages respond more fully to the user's request, and their content differs significantly from the canonical ones. If you are sure that such pages won't be useful in search, prohibit indexing in the [robots.txt](https://yandex.com/support/webmaster/en/controlling-robot/robots-txt.md) file.
- The canonical URL is not accessible to the robot — it redirects to another page or is [closed from indexing](https://yandex.com/support/webmaster/en/adding-site/indexing-prohibition.md). This means it [can't be included in the search](https://yandex.com/support/webmaster/en/site-indexing/excluded-pages.md). In this case, a non-canonical URL can be included in the search instead of the canonical URL, provided the robot can access it.
- The canonical URL points to another domain or subdomain.
- Several canonical URLs are specified.
- A chain of canonical URLs is specified. For example, for `example.ru/1`, the canonical URL is `example.ru/2`. At the same time, `example.ru/2` has the canonical URL `example.ru/3`.
## Questions and answers {#qanda}
{% cut "The rel="canonical" attribute points to the page it's located on. Is this an error?" %}
No. If the `rel="canonical"` attribute on the page refers to this page, the robot considers it canonical.
{% endcut %}
{% cut "How do I reinclude a non-canonical page in searches?" %}
If a page was excluded from search results for being non-canonical, it means that the robot found the `rel="canonical"` attribute with the canonical URL in its HTML code or HTTP header. Delete this reference and check that the page you want to include back in the search [is not closed to indexing](https://yandex.com/support/webmaster/en/adding-site/indexing-prohibition.md).
{% endcut %}
{% cut "Contact support" %}
If you still have questions about using the `rel="canonical"` attribute, provide the examples of pages that you are having problems with in the form below.