Those of you who keep on top of the Google Webmaster blog will have no doubt read about their recent pagination fix, or at least their advice on how to deal with it and become more Google friendly.
What is pagination
For those of you who aren’t sure what pagination is, in layman terms it’s a list of web data more commonly seen on e-commerce sites to display multiple products, like here, however it can be used for many different reasons including:
Most forums have tons of replies to threads, from a user point of view you want to break these up (pagination), however Google would prefer all these replies or comments to be on the same page.
Most blogs list a certain number of posts on the home page and then use pagination to allow the user to flick through historic posts, again it saves having a page 10 feet long, but Google don’t like having to trail through all the paginated pages.
Most e-commerce sites make use of pagination on product pages, if you have 900 products in a particular category, you hardly want to list them all on one page? So you use pagination to make it more user friendly, however Google is only ever going to crawl so much of the pagination.
Why is it an issue?
OK, we all understand what it is and why it is used, however what you may not realise is that it can cause huge issues from an SEO perspective. Think about it, if one of your category/tread/blog pages has 25 pages of pagination for Google to crawl through how far do you think they are going to crawl? Well, probably all the way, however why would Google index pages that are 25 clicks away from the top level?
This is the issue, helping Google understand that these paginated pages are important, insomuch that they contain valuable content and links, in effect they are the same page.
The noindex and canonical mistakes
Historically pagination has never been dealt with very well, in particularly the use of the rel=canonical tag is a big no no… Webmasters would use the canonical to link all the paginated pages back to the first page of the pagination, basically telling Google this is the only page that holds any importance. So what about all the content/links on the paginated pages, do you not want these indexing and following?
Using the rel=canonical to sort this is not the way and will mean a large quantity of your site does not get indexed.
Some webmasters also opt to no index everything but the first paginated page to avoid duplicate content issues, however it has the same sort of problems in that Google will not index all the pages linked to within the pagination.
So we want Google to index all the links and content in the paginated section of the site but Google don’t want to index every page to get this done, Google have now offered a little advice on the subject.
Googles solution to pagination woe!
Recently Google have come up with some recommendations around pagination and I just want to spend a little bit of time looking at them and how we implement them.
Rel=next / Rel=prev
Much like the rel=canonical we can use the rel=next and rel=prev to let Google know about pagination elements on your site and to indicate relationships between the pages. So to page 1 in the pagination you would add the rel=next + the next URL, all other pages in the pagination will need to implement next and prev along with the previous and subsequent URL (read more here Google Pagination)
Adding these tags in has a couple of effects:
A) It let’s Google know that all the content and links can be attributed to the first page and not spread out along 25 pages
B) Helps Google to understand what the most useful page is, where to send users and ultimately which pages to index.
The View all Solution
According to Google having a view all page that clumps all your pagination together is a better user experience, not so sure personally but lets trust them on this occasion. However, apparently there is another reason to implement your view all page and that’s because this is really what Google wants to index. So the recommendation is as follows;
a) Use the next/prev tags to indicate the relationship between paginated pages
b) Create a view all page as this is what Google really wants to index
I would also recommend using the rel=canonical on the first page in the pagination and pointing it to the view all page, this will prevent you confusing Google and getting mixed results across all your pagination.
How Can it Affect Quality Signals?
Google have stated that they will carry on trying to index pagination and deliver the best result to searchers regardless of the above advice. However, my feelings are that you need to follow the above advice in order to increase the quality of your site structure. We all know that Panda was rolled out to squash sites with low quality content, however from the examples I have seen this doesn’t just relate to the actual text but also relates to how pages/content is structured.
Pagination is popular among e-commerce websites, so is having duplicate product descriptions that are being used across the web. In order to stay out of trouble you need to provide Google with as much information as possible and take on any little bits of advice that are thrown your way, make use of micro formats, get listed in Google places and implement the above pagination advice, I personally think reacting to these changes is a sign that your website is genuine and therefore puts you out of harms way.