Conflicting advice for link rel=canonical on paginated results page

[bernardjhuang]

I have a website that displays data about businesses in a directory style format (e.g. Yelp). Most of the pages have paginated search results because of the # of listings in my directory.

From this blog post (https://moz.com/blog/pagination-best-practices-for-seo-user-experience) by Rand Fishkin, he gives examples of how he would implement canonical, title, and description tags for paginated results pages:

Whatever you do, DO NOT:

Put a rel=canonical directive on paginated results pointing back to the top page in an attempt to flow link juice to that URL. You'll either misdirect the engines into thinking you have only a single page of results or convince them that your directives aren't worth following (as they find clearly unique content on those pages).

Add nofollow to the paginated links on the results pages. This tells the engines not to flow link juice/votes/authority down into the results pages that desperately need those votes to help them get indexed and pass value to the deeper pages.

Top Page Title: Theatres & Playhouses in Princeton, New Jersey
Top Page Meta Description: Listings of 368 theatres, playhouses and performance venues in the Princeton, NJ region (including surrounding cities).

Page 4 Title: Page 4 of 7 for Princeton, New Jersey Theatres & Playhouses
Page 4 Meta Description: Listings 201-250 (out of 368) theatres, playhouses and performance venues in the Princeton, NJ region (including surrounding cities).

Alternate Page 4 Title: Results Page 4/7 for Princeton, New Jersey Theatres & Playhouses
Alternate Page 4: Description: -

Which is conflicting advice that I've recently received from an SEO firm that I've been working with who told me to put a link rel=canonical directive on paginated results pointing back to my top page.

Also conflicting is Yelp's implementation of paginated search results. A look at the 2nd page of a search query gets:

<meta name="robots" content="noindex">
<link rel="prev" href="http://www.yelp.com/search?find_desc=Best+Coffee+Shop&amp;find_loc=San+Francisco%2C+CA" />
<link rel="next" href="http://www.yelp.com/search?find_desc=Best+Coffee+Shop&amp;find_loc=San+Francisco%2C+CA&amp;start=20" />

with no change in their <title> tag and the swapping of listings in their <meta description> tag.

My question is what implementation method have you guys seen to work the best for paginated results?

[aakk9999]

Welcome to WebmasterWorld, bernardjhuang :)

... SEO firm that I've been working with who told me to put a link rel=canonical directive on paginated results pointing back to my top page

This advice is wrong and is in fact listed as Mistake #1 in Google's blog post discussing five most common mistakes with rel canonical:

Mistake 1: rel=canonical to the first page of a paginated series
Specifying a rel=canonical from page 2 (or any later page) to page 1 is not correct use of rel=canonical, as these are not duplicate pages. Using rel=canonical in this instance would result in the content on pages 2 and beyond not being indexed at all.
(...)
In cases of paginated content, we recommend either a rel=canonical from component pages to a single-page version of the article, or to use rel=”prev” and rel=”next” pagination markup.

5 common mistakes with rel=canonical [googlewebmastercentral.blogspot.co.uk]

Regarding Rand Fishkin's advice - this was fine before the introduction of rel=prev/next. Now the rel=prev/next is the best practice so I suggest that you use Yelp's implementation.

[dipper]

Rand is correct - from Google themselves .. [googlewebmastercentral.blogspot.com.au...]

[bernardjhuang]

Thanks for the warm welcome aakk9999 and dipper!

Regarding Rand Fishkin's advice - this was fine before the introduction of rel=prev/next. Now the rel=prev/next is the best practice so I suggest that you use Yelp's implementation.

This makes the most sense to me.

[FranticFish]

Has anyone here tested rel=prev/next in isolation? My experience is that it is not enough on its own.

I have been advising on tech optimisation on a site recently which had an AJAX pseudo pagination class (i.e. a 'more articles' link) that kept loading content into the page. This is non-spiderable.

So, in addition, they were using a famous SEO plugin which was generating rel=next/prev tags correctly to their paginated pages e.g. /category/page2, /category/page3 etc.

No pages beyond the first root /category/ page were indexed.

[dipper]

was there a rel canonical tag in place which showed only the first page in the pagination sequence regardless of which paginated page you were on?

[FranticFish]

No. The SEO plugin adds one to every page but they all point correctly to that page rather than the root page.