Common rel=”canonical” errors in video SEO

Do you have two pages on your site that are practically identical yet necessary in order for your site to work? And do you need one of them to be indexed in Google search? Then you should use the rel=”canonical” tag. It’s a useful tool in any SEOs arsenal. But where can you find the tag? It sits in the <head> of the duplicate page, and points to the version you want to be indexed. In other words: It tells Google and other search engine crawlers that this version is the “canonical”.

With video marketing, there are common technical implementations that require specific use of the rel=”canonical” tag. For example, to prevent the indexation of duplicate and low-quality pages. There are also common ways many plugins and tools default to implementing rel=”canonical” that can hamper video SEO.

In this post, we’ll break down the common mistakes typically made with rel=”canonical” for video SEO. And we’ll explain how to avoid them!

1: Leaving isolated video hosting pages indexable with self-referencing canonical tags

When self-hosting videos use a custom player and CDN, it’s very common to embed the videos as an encapsulated media file on an otherwise blank page. A video player can then reference the files via iframe or JavaScript. These pages will often live on a subdirectory or subdomain, e.g. videos.example.com/video-5.html. They usually serve no functional purpose for users, except to offer a location from which video players can find and pull in video files.

Because the purpose of these pages is technical and they’re often not linked to (beyond instances in the video player code), the pages are true duplicates. Which means it’s not appropriate to leave them open to the index. However, the pages do need to be crawlable. Otherwise, Googlebot Video won’t be able to find the video files and index the videos.

The best solution is therefore to implement rel=”canonical”. This will make your pages crawlable, while telling Google not to index them. Instead, Google will see them as subsidiary assets for the core page that the videos are presented on. This could mean that you need to adjust default self-referencing rel=”canonical” rules, so these isolated page include self-referencing canonical tags by default.

Solution: Use rel=”canonical” to point isolated video pages to the pages they are primarily embedded on.

2. Implementing canonical tags for the same video on multiple pages

When indexing videos, Google considers a video as a child of a page rather than an individual asset with its own distinct URL, as they do with image files. What does this mean? It means duplication at a video level currently isn’t a particular consideration for Googlebot. Even if you include the same video on two pages and implement structured data so Googlebot can find and index the videos, Google doesn’t per se consider them to be the same video. 

Note: This doesn’t mean you can publish the same video on as many pages as you like with no consequences. If the pages are otherwise very similar, target similar keywords, and have identical video titles and thumbnails, then this can still create confusion and ranking cannibalization. However, this would be a problem at page level rather than video level.

In the future, there may well emerge a tool to indicate duplication at a media asset level. For now, though, the best solution is to ensure that each page you create contains unique content in addition to any duplicated assets. And trust Google to know which page is most appropriate to rank for any given query.

Solution: Don’t use rel=”canonical”. Just make sure every page has unique copy and other unique media.

What if you need to use the same video multiple times?

For example, if you want to use it on a help page and a blog post. Rest assured, you can still do this without worrying about duplication. However, some people still feel that you should use rel=”canonical” to indicate this duplication. Their reason? The videos are identical, so one of the pages has to be more appropriate than the other to rank. This isn’t true, though. Because rel=”canonical” only operates at a page level rather than a media asset level, it would be an incorrect use of the protocol.

Read more: How to create a video website »

3. Using URL parameters to indicate timestamps without rel=”canonical”

If you want to stipulate to a browser to start a video at a specific point (rather than the start), you would probably use URL parameters. For example, YouTube videos work with a “?t=” parameter, and Wistia videos work with “?wtime=”. 

URL parameters are a really useful feature. But they can be especially beneficial for Video SEO, since they allow you to create URLs to use in Clip Schema in combination with VideoObject. This, in turn, allows you to ensure that your videos are indexed for “Key Moments” in Google search, thereby taking up more attention and space in the results pages for any given query.

However, if you use URL parameters like this, you’re technically creating duplicate pages for each “clip”. Which Google isn’t always able to assume should be defaulted to the page root. You therefore need to implement rules which ensure that any URL using the query parameter that stipulates video rules is automatically canonicalized back to the page without this query parameter (typically the root URL).

Solution: Ensure that canonical rules automatically add a tag for URLs with a timestamp parameter. Make it point to the variation of the URL without this parameter included.

4. Automatically canonicalizing query parameters that determine lightboxes and video galleries

Do you have a video lightbox or gallery on your site? If so, then chances are you’re using video plugins or embed codes that use query parameters or hashes to indicate a URL variation.

In such cases, the video player won’t load until the JavaScript is triggered, which is indicated by the adapted URL. For Googlebot to be able to find, render, and index these embedded videos, the URL variations themselves must be crawlable and indexable. This might sound contradictory to the advice we gave before. But there’s a crucial difference between the two situations: The change in content on the page, which occurs with the JavaScript function.

You need to make these types of dynamically generated pages, such as those created through Wistia Channels, available for search engine crawlers. However, we recommend that you don’t use noindex or rel=”canonical” tags. This does mean you have to carefully consider the automatic canonical rules. In addition, you have to ensure that any parameters you use are not automatically grouped in with timestamp parameters, or tracking parameters that are implemented by other analytics platforms. These should be automatically canonicalized.

Solution: Don’t add canonical tags to these pages.

In summary

The rel=”canonical” tag can be a useful tool in your SEOs arsenal. It specifies the “canonical URL,” or the “preferred” version of a web page. This means you can avoid pointing Google towards duplicate content. So if you use the rel=”canonical” tag correctly, it will improve your site’s SEO. Awesome, right? That’s why it’s good to know what the common mistakes are when using this task, and how to correct them.

Keep reading: How to use the new video indexing report in Google Search Console »

Coming up next!