You’ve surely heard that pagination can be detrimental to a website's SEO: this is both true and false, because the problem is not with pagination itself, but with the way it is implemented.
Indeed, the pagination poses the problem of crawl and indexing because the more paginated content you have, the more pages the Google robot must visit to successfully index all your content, which jeopardises the optimization of your content. There are 2 conclusions:
- Do not implement pagination if you do not need it: sites with very few products for example.
- If you need it, keep reading to get the detailed method!
Pagination concerns all websites, from the small amateur blog to large e-commerce sites with several thousands of pages. It is a technique for organizing content into several dynamic pages.
On an e-commerce site, it usually appears on the category pages. On a news site, articles are often split into several pages or presented as slideshows. On forums, groups and topics are usually spread over at least 2 to 3 pages. Even chronologically organized blogs need to implement page management. This is a much more complex subject than it seems, and it makes all SEOs, whether they are novices or experts, turn their heads. There is really no best pagination. Each site must implement the one that is the most relevant to its internal linking and performance.
With the arrival of the infinite scroll which allows to have all the content on a single page (at least in appearance), one can indeed wonder why set up a pagination. In fact, it is essential from a user experiencepoint of view. It allows you to prioritise your products or articles, to make navigation easier. For some publishers, it is also an opportunity to multiply the number of displays of advertising campaigns. With a clean code and a good structure, it also constitutes a "path" leading robots to deeper pages. Finally, it limits the server load which is often the cause of slow running high traffic sites.
But the implementation of a pagination involves risks.
Waste of GoogleBot crawls
Once your pagination is put into production, Google will crawl all the pages and may get bogged down at the expense of important pages.
SEO juice dilution
Incorrect code implementation can dilute SEO juicethrough numbered pages that are often less relevant.
Paginated pages can be a source of duplicate content if they are not managed properly, especially when facets are involved.
Low quality content
The majority of paginated pages contain only the bare minimum of content. Beware, Panda is hunting bad pages down!
There are several pagination systems.
The numbered pagination combined with a "view all" page
If your paginated series has an alternative version contained on a page, Google will try to detect it and index it. This system requires a canonical tag (with the "view all" page as the canonical URL) on all pages. This will guide the robots to the page more easily.
This is certainly the easiest option to implement technically and the most user-friendly. It also avoids duplicate content thanks to the canonical tag. Nevertheless, it is not suitable for all types of sites. Large e-commerce sites cannot opt for this solution because it would generate endless pages that take a long time to load.
Sign-post pagination in the HTML code
If the first option is not possible, you can use the standard tags rel= "next" and rel= "prev ". Google takes into account these tags to index pages. It now knows that paginated pages are not individual pages and rarely indexes them. In practice, paginated pages must have a canonical URL that includes their own page number.
For example, the paginated page www.example.fr/guide-seo/4 must have the following canonical tag :
www.exemple.fr/guide-seo/4 « />
This method is suitable for sites with long load times or sites that do not have a "view all" page. On the other hand, it limits the control over page indexing. You could end up with page 2 indexed, when it is page 1 that is actually relevant.
The infinite scroll
The noindex follow
The noindex follow prevents the indexation of paginated pages, while continuing to distribute SEO juice to the articles or products they contain. This trick allows you to prioritize the indexing of the first page, but beware, it does not prevent the robots from crawling further pages. It can therefore not be used for PageRank sculpting. If your secondary pages are of real interest, you can authorize their indexing.
Configuring URL parameters in the GSC
Configuring the URL parameters in the Search Console allows you to tell Google how it should behave when dealing with the different types of pages listed. It is therefore possible to indicate whether it should explore or not the numbered pages. But be careful, robots will not always listen to you!
Reduce the number of pages accessible from the page numbering
It is useless and counterproductive to integrate all the page numbers. The best practice is still to propose the first and last page, with "next" and "previous" links.
A common mistake is to integrate a canonical tag including the first page as canonical URL. This technique was once used to condense all the authority of the pages on the first page. Google now advises against it in its pagination guideline.
“Paginated” pages that should not exist
This is a server response issue: some paginated pages respond with an empty list and an HTTP 200. For example, a list has 3 pages, but www.mysite.com/list.html?p=4 responds with an empty page, instead of returning HTTP 404.
Use noindex nofollow
Some sites think they are optimizing their crawl budget by asking Google not to follow pagination links. It is a technique that is debatable, but the SEO juice will not be sent to the secondary pages. So it is better to have a very strong internal linking instead!
Pagination is therefore a major issue in on-site optimization. The multiple solutions proposed by Google have advantages but also disadvantages. It is therefore advisable to measure the relevance of its pagination using a crawler and if possible a log analyzer.