Sep 16, 2011 121 reads by Navneet Kaushal

The third day of the SMX New York 2011 kicked off with the session titled- Pagination & SEO. It was focused on achieving effective indexing and ranking by implementing complex, paginated sites. The rel=canonical and noindex tags were the center of discussion here. The panel constituted of-


  • Vanessa Fox, Contributing Editor, Search Engine Land

Q&A Moderator:

  • Simon Heseltine, Director, AOL Inc.


  • Richard Chavez, Sr. Director, SEO, PM Digital
  • Vanessa Fox, Contributing Editor, Search Engine Land
  • Maile Ohye, Senior Developer Programs Engineer, Google Inc.

Richard Chavez took the floor first and started off with a question- “Why merchandise your products in multiple categories?” the answer is that consumers look for things by taking different routes, either by category or need etc. This is the thought process that Richard took to explain further that if one targets various target points for consumers- or evaluate different entry ways- one can get more customers.

But this miracle solution has some challenges too. And that too technical in nature-

  • You can end up with duplicate content for product URLs
  • If you choose to do pagination for each merchandising option; then again you can get trapped in the duplicate content hole.

Richard then explained with the help of rankings reports in various search engines that the best option for you is to make the right start. In other words build it right from the first time- This means you must have unique URLs for each product leaving aside the leading page.

He recommended paginating at the category level and canonical to the root page. Also for better indexing you must include product URLS in XML sitemaps. This is the steps he suggested for working with existing sites:

  • Rel = canonical in those places
  • XML Sitemap feeds for one product URL
  • Point to the main page in pagination
  • He recommended pagination via AJAX
  • If using AJAX do provide an XML feed to Google and also use the Google AJAX specification.

The Takeaways:
The beginning has to right. Build it right the first time and with existing sites use canonical tags.

The next speaker to take the stage was Maile Ohye, Senior Developer Programs Engineer, Google Inc. Maile starts off by saying that pagination has been an ongoing issue. They have been approached by webmasters to come up with a solution. And she will now introduce what Google has newly come up with.

She put forward Google's point of view that the rel=”canonical” attribute was not meant to cluster multiple pages (articles, product lists, etc.) to page one of that series. However, it can be used to cluster multiple pages to a “view all” page.

She also explored the Side effects of pagination-

  • Links diluted into several pages (issues with indexing)
  • Chances are the the Relevant page of the series may not be reflected in search results
  • The “View All” on the page might make it difficult to load

Maile said that searchers liked the view all page. They want all information in one go and they like it for longer and better if the “view all” page doesn't take long to load. This means that – Google will serve searchers the "view all" page by default. Provided your site gives the signal to Google that it is paginated and there is a view all. The benefit- the indexing properties, such as links, will all be passed automatically.

The next option is if you don't want a view all page ranked, then go for:

  • rel=”next”/rel=”prev” option
  • You can block it with robots.txt or meta noindex.
  • You can use rel=”next” and rel=”prev” to cluster all of the component pages into a single series. This will lead to consolidation of all the components in the series and the most relevant page in the series will rank for each query.

Pagination And SEO

She said that, “you must declare in the head section of the page, page two to the second to last page should be doubly linked with both rel next and prev. If you do not have a view all page, you can implement this. If you have the view all, you can do both.”

Now the big question- Why use rel next and rel prev instead of rel canonical to page one?

She answered with an explanation that Rel canonical is for duplicate content and paginated results are not always duplicate content. She further advised that in the case of session ID URLs you must use rel canonical. The thing to keep in mind is that rel canonical only indexes content from the canonical URL. This means that as the other pages have different content, the engines won't pick it up using rel canonical. So here is the time for- rel next and previous come in. She said, “It is like a series and sequence of content, which link all the content into one.”

She further elaborated on this by saying that component URLs in the series should agree with pre/next values. It is a separate thing form rel canonical and can be used with it or without it.

The session then went to Q and A and she then answered the questions from the attendees. On being asked the ideal load time for a page, she said that 2-3 seconds is probably a good benchmark but there is no one number.

The Takeaways from the session and Q and A
Each page on the paginated results must have unique title tags.Google will judge on the basis of that only. And it may return page 3 if the title tag makes sense, however most likely, it will be page one that will be shown in search results.

Pagination & SEO: SMX New York 2011, Day 3!, 5.0 out of 5 based on 3 ratings
Navneet Kaushal

Navneet Kaushal

Navneet Kaushal is the founder and CEO of PageTraffic, an SEO Agency in India with offices in Chicago, Mumbai and London. A leading search strategist, Navneet helps clients maintain an edge in search engines and the online media. Navneet's expertise has established PageTraffic as one of the most awarded and successful search marketing agencies.
Navneet Kaushal
Navneet Kaushal
Most popular Posts
Upcoming Events
Events are coming soon, stay tuned!More