As I did last year with my Long Tail Link Building presentation that I gave at SMX Israel, I’m not going to just embed the presentation, I’ll try to take you throw the presentation so you can get the most out of it.
Before starting, what is pagination?
Pagination is the process of dividing (content) into discrete pages, either electronic pages or printed pages. (From wikipedia)
So, here is the presentation:
Let’s hop in to it slide by slide
What is the problem with paginating pages on the internet?
I don’t think I need to add anything to the slide
Shows the most popular places on the web were we find pagination (Articles, forums and ecommerce sites).
We have 3 different ways that evolved over time, dealing with pagination:
- meta robots tag
- canonical tag
- next and prev tags
In the upcoming slides we will see the way the tags effect the visibility of the pages in the search engines. In all examples I use an article divided in to 3 pages.
The first option is to make sure the only page that will appear in the search engines will be the first page. This is done by using meta robots tag telling the crawlers they can crawl the page but not index it. This tag is added to the second and third page.
The visual result of the use of the meta robots tag noindex follow used with pagination. The search engines will only really know about the first page even though we let them crawl all three pages. In result only the first page will appear in search results.
How does this affect the SEO strategy?
Links that will go to the second or third page will not help rankings for those pages or for the first page of the article, and you end up losing some anchor text and maybe some link power, probably you will get the trust but that is just a guess.
The second option is adding a canonical tag to pages two and three. Note that Google says that the canonical tag acts as a strong hint for duplicate content. This means that Google may decide not to pay attention to the canonical tag especially that we are not using it on pages with similar content. From my experience, I don’t remember ever seeing a page with a canonical tag pointing to a different page in the rankings, even when it was defiantly not duplicate or even similar content.
The visual result of using the canonical tag for pagination. The search engines will be able to crawl all three pages but most probably will show only the first page in search results.
In terms of inbound SEO strategy, we are better off with the canonical tag because all the power will be flowing to the first page. This is much better than losing some power in the previous option.
The third and preferred option is using rel=“prev” and rel=“next” tags. Notice how there is a chain that connects all the paginated pages together. I must add that even though these tags are in the HTML4 protocol (see 12.1.2 Other link relationships at http://www.w3.org/TR/html401/struct/links.html) only Google uses them as a pagination signal.
The visual result of using the rel=“prev” and rel=“next” tags for pagination. Crawlers will crawl and index all three pages and they all can show up in the search results.
The best part about this is that Google knows now that all three pages are actually only one article and will treat it like that too. So any links coming in are actually pointing to all three pages so you are not losing any link power or anchor text. Based on your quarry Google will return the most relevant part of the article.
There is a forth option which I don’t deal with in this presentation and that is to load the entire article on one page and use JS to go throw the different sections. This is still not as good as using the rel=“prev” and rel=“next” tags because with the JS you will always be sent to the first section and with the rel=“prev” and rel=“next” tags you might be sent to a different page.
Pagination with canonical tag. Google says “User testing has taught us that searchers much prefer the view-all, single-page version of content over a component page containing only a portion of the same information with arbitrary page breaks”
So how do we deal with a situation where I have paginating pages and a view all?
Google also supplies the solution: if the view all page lodes fast (less than 3 seconds) use that as the default page and have the paginating pages (of course they already have rel=“prev” and rel=“next” tags) point to the view all page with a canonical tag. If you view all page takes a long time to load then you should prefer the paginating pages for best user experience that means have a canonical tag from the view all page pointing to the first page in the series.
Pagination with parameters in the url and a canonical tag, as you can understand that this is starting to get complicated.
The first example that we have is in a category section of a website that is also being sorted by popularity of the items. I’ll add that this is also the default view of the category page. This is why the canonical tag is pointing to the same url just without the parameters. The next step is to add the rel=“prev” and rel=“next” tags but we got a problem, because we need to add in to the tags the parameters in the url this will create 2 chains with the exact same content (remember we said that the default view is by popularity) so this means it is unnecessary to use the pagination tags for this parameter.
The second example is a subcategory that is sorted by price. The default view is by popularity just like in the first example, so the canonical tag is pointing to the url without parameters. Unlike the previous example here we want the search engines to know that we have other ways of sorting the products, so that if a user will search for my category with the word cheap I hope to have the first page sorted by price ranking.
The rules you need to remember when adding rel=“prev” and rel=“next” tags to your paginating pages.