What is a canonical tag?
Canonical tags indicate the master version of a page. If the page itself is the master, the canonical tag can be self-referential. However, if the page is similar to (or a duplicate of) a master page, the canonical tag should point to the master version.
By indicating the master version of pages to search engines, canonical tags allow you to recommend the version you would like indexed. Just remember, by canonicalising elsewhere you’re telling search engines that there is another version of the page more suited to indexing, effectively telling them not to index the canonicalised page. In this respect, a canonical tag makes a page non-indexable.
Canonical tags pass link equity to the page they reference. This makes them useful when used on parameterised pages which may be linked from external sites. This is also particularly important when there are multiple parameterised versions of a URL, all of which (when canonicalised) will pass link equity to be consolidated on a master version.
Where should you use canonical tags?
A canonical tag should be marked up in the head of the page as:
<link rel=”canonical” href=”https://example.com” />
It’s good practice to implement canonical tags on all pages. Where a page is the master version, it should contain a self-referential canonical, so it explicitly declares itself to search engines as the master. Where pages are duplicate (or similar enough to be considered duplicate), they should canonicalise to the master version.
Cross-domain canonical tags
A cross-domain canonical tag is one which points to an external domain. They are useful when dealing with associated domains or guest articles. They function the same way as a normal canonical (by hinting to search engines that the master version is the one you want indexed), however search engines are more inclined to ignore cross-domain canonicals if the canonicalised page has more authority.
- Google SEO Ranking Factors for 2021
- How to Clean Up Your Link Profiles
- How to Create Blog Posts That Rank High in Google
- Link Building Strategies for New Websites 2021
- Introduction to Blackhat and Whitehat SEO
- How to Run an Initial SEO Audit on Your Website
- How to Deal temporarily Out of stock & seasonal products pages
- How can we better understand what local results Google are serving to visitors?
Why are duplicate pages an issue?
Canonical tags help deal with the issues caused by duplicate pages. However, it may not always be clear what issues duplicate pages pose to sites. Outlined below are some duplicate pages issues that could be solved with canonical tags.
Keyword cannibalisation occurs when pages on the same domain fight to rank for the same query. This can potentially impact the ranking of both pages due to split authority.
If it’s important for both pages to be accessible (one may be a parameterised version, for example) canonical tags can be used to pass link equity to the master version, reducing the issue of split authority.
Canonical tag problems
Incorrect canonical tag usage
It’s surprisingly easy to misuse canonical tags. Canonical tags must always point to a master version, (even if the master version is itself), and they’ll be considered invalid if they point to a non-indexable URL. Think a URL with a non-200 status code or a URL with a noindex tag.
A further misuse of canonical tags is often seen in paginated series, where page 2 onwards canonicalise down to page 1 of the series. Unless canonicalising to a View All page, pages in a series are expected to stand on their own and should contain a self-referential canonical tag.
Non-optimal canonical tags
Canonical and redirect chains
Canonicalising to a URL which then canonicalises or redirects elsewhere mixes the signals sent to search engines. While they may follow the chain and may understand the intended effect, mixing signals reduces the likelihood that search engines will adhere to canonical tags.
Canonicalised pages in sitemaps
It is not uncommon to find canonicalised pages in sitemaps, however this isn’t ideal as it confuses the signals sent to search engines. A sitemap should be a list of URLs deemed important for indexing, while a canonical tag tells search engines not to index the page.
If a sitemap has canonicalised URLs, we recommend auditing the sitemap and replacing those URLs with their respective master versions.
Mixing canonical and noindex tags
Implementing both canonical and noindex tags on the same page is not recommended. A canonical tag implies to Google that any value should be passed to the referenced page and that page should be indexed in its place, whereas a noindex tag implies the page does not have value and should be kept from the index.
The problem with canonical tags
Canonical tags are great for informing search engines of the relationship between master and duplicate URLs, however they’re only a hint, not a directive. This means search engines could choose to ignore them.
If a site has a lot of confusing canonical tags, (like canonicals pointing to other non-indexable pages or pages in a series canonicalising down to page 1), search engines may begin ignoring the canonicals and picking their own instead.
Why are canonical tags important for SEO?
Though canonicals are a useful tool in your arsenal, they may not always be the most effective choice. As with most things in SEO, canonical tag usage is subjective and situational. Below are a few common considerations you should make before deciding on using a canonical tag.
Would a redirect be better?
While canonical tags do pass authority, sometimes it’s more appropriate to use a redirect. Imagine you have duplicate domains, such as a www and non-www version of your domain. Some sites may try to use canonical tags to handle this duplication, however this is not recommended since the canonical tags could be ignored by search engines if the other signals (such as internal linking and backlinks) contradict them.
A good rule of thumb for deciding whether a duplicate page requires a redirect or a canonical tag is to think about whether the page needs to be accessible. If it does (like a parameterised page would) then a canonical tag should be your choice. If it doesn’t (in the case of www vs non-www) a redirect is favourable.
How important is it to keep this page out of the index?
If it is absolutely vital to keep a page out of the index, (or if a page is in the index and needs to be removed), a noindex tag might be favourable over a canonical tag. This is because a noindex tag is a directive and will be adhered to immediately once crawled by Google, whereas a canonical tag is a hint and can be ignored.
Is the page actually duplicate?
While it might be appropriate to use canonical tags on near-duplicate pages (and in some cases is recommended), use this method with caution. If search engines feel the canonicalised page is not similar enough to the page declared as the master, they will ignore the canonical. If this problem is widespread, it can damage search engines’ trust in all your canonicals.