Pagination using rel="next" and rel="prev"

[engine]

Pagination using rel="next" and rel="prev" [googlewebmastercentral.blogspot.com]

Much like rel=”canonical” acts a strong hint for duplicate content, you can now use the HTML link elements rel=”next” and rel=”prev” to indicate the relationship between component URLs in a paginated series. Throughout the web, a paginated series of content may take many shapes—it can be an article divided into several component pages, or a product category with items spread across several pages, or a forum thread divided into a sequence of URLs. Now, if you choose to include rel=”next” and rel=”prev” markup on the component pages within a series, you’re giving Google a strong hint that you’d like us to:

Consolidate indexing properties, such as links, from the component pages/URLs to the series as a whole (i.e., links should not remain dispersed between page-1.html, page-2.html, etc., but be grouped with the sequence).
Send users to the most relevant page/URL—typically the first page of the series.

[1script]

Send users to the most relevant page/URL—typically the first page of the series.

It may be typical but it's not always true. I find it that on forums specifically, the discussion may actually get to the most interesting parts only on page 2 or even further inside, and even without tags that essentially kill off all pages except for the first one, the consequent pages have trouble ranking. I guess, mostly because most internal links are pointing to page 1.

In other words, at least on forums, page 1 may not be the best page you want ranking even on keywords in the title of the discussion so using rel="next" and rel="prev" will just further exacerbate the issue.

[martaay]

What are peoples thoughts on using this feature on a business directory, or is this really designed for articles with split content?

[1script]

or is this really designed for articles with split content?

I honestly don't know what the original intent was but now it's used in a context that's different from what Google wants to use it for. WordPress, for example, by default adds rel=”next” and rel=”prev” to articles that have nothing to do with one another except that they were published sequentially. This is an example of a wrong signal Google is trying to pick up. The fact that different URLs are "linked" through the chain of rel=”next” and rel=”prev” directives does not mean that whichever is the last one in the chain, the most previous, so to speak, should rank best (apparently Google now thinks it's Page #1). In WordPress interpretation it is simply the oldest article.

Anyhow, if Google will start interpreting these directives as pagination signals, many WordPress installs, at least the most recent version 3.2.1 are in danger.

[BrodyDodes]

A site I work with currently uses rel=canonical to point pages 2 and on back to page 1. This isn't recommended but prevents all the paginated pages from being in the search index. Any thought if rel=next/prev will account for this and only index page 1? Or will noindex tags be required on all subsequent pages?

[deadsea]

I like what Google is trying to do here. Seems like there is a lot of details and corner cases for them to work through. I was wondering why I'd been seeing 404 page-all requests from Googlebot in my logs.

[1script]

A site I work with currently uses rel=canonical to point pages 2 and on back to page 1. This isn't recommended but prevents all the paginated pages from being in the search index.

View-All and Page 1 are quite different unless your Page 1 contains all the content. In which case why would there be other pages?
Presumably, all those pages also have content or else why should there be additional pages, so I just don't see any reason to do rel="canonical" to Page 1 if content of Page 2+ is different.

In other words, why would you make them only index page 1 if other pages can also stand of their own and hopefully rank as well?

[rlange]

1script wrote:
I honestly don't know what the original intent was but now it's used in a context that's different from what Google wants to use it for. WordPress, for example, by default adds rel=�next� and rel=�prev� to articles that have nothing to do with one another except that they were published sequentially. This is an example of a wrong signal Google is trying to pick up. The fact that different URLs are "linked" through the chain of rel=�next� and rel=�prev� directives does not mean that whichever is the last one in the chain, the most previous, so to speak, should rank best (apparently Google now thinks it's Page #1). In WordPress interpretation it is simply the oldest article.

Anyhow, if Google will start interpreting these directives as pagination signals, many WordPress installs, at least the most recent version 3.2.1 are in danger.

That's an obvious problem. So obvious, in fact, that I have to assume there are other factors that Google looks at when determining if

rel="prev"

and

rel="next"

are indicating paginated content as opposed to a basic list. As the blog post says, it's just a hint, not a directive.

--
Ryan

[tedster]

Any thought if rel=next/prev will... only index page 1?

I thought the blog post hinted strongly at that possibility too. I surely hope not. I've been using rel="next" for a long time on one site and the search traffic to those inner pages has been excellent. I'd hate to see it go away... or end up going to Page 1 where the search keyword doesn't exist.

[mark_roach]

I have just implemented this on the category pages on my site and am hopeful it will work well for me.

Page 1 is typically the strongest page with respect to internal and external links and in contrast Page 20 is typically not externally linked nor well linked internally.

Because of this even though the search term (typically location in my case) the searcher is looking for is on page 20 that page may not might have the power to rank. If the same search term appeared on page 1 then the page would be returned in the search results.

Consolidate indexing properties, such as links, from the component pages/URLs to the series as a whole (i.e., links should not remain dispersed between page-1.html, page-2.html, etc., but be grouped with the sequence).

I am hopefull that the above means that each page in the sequence will carry the same weight with respect to links. This would then mean that whether the data is contained on Page 1 or Page 20 your site (sequence) is equally likely to be returned.

Send users to the most relevant page/URL—typically the first page of the series.

So your sequence is chosen because of the contents of page 20, however page 1 may be returned in the results.

If this is how it works I am sure google will test by sometimes returning page 1 and at others page 20 and monitoring bounce rates etc. before deciding which page is the most appropriate for a sequence.

Interestingly, or perhaps unfortunately because it muddies the water, the day (saturday) I added this to the site (so before the majority of my pages had been re-spidered) I saw a five-fold increase from Google to my "view all" pages and the next day the "view all" visits were 10x normal levels. Yesterday they were at 3x normal, so it will be interesting to see if they return to normal tomorrow.