Best Way To Handle Pagination?

[seoskunk]

Hi I was wondering what you guys thought was the best way to handle pagination for Google

EXAMPLE 1
Category with rel = next and rel = previous links in header
INDEX,FOLLOW
Canonical Url set for every page
ie. page 2 would have an canonical url of "http://www.example.com/category/?page=2"

EXAMPLE 2
Category with rel = next and rel = previous links in header
INDEX,FOLLOW
Canonical Url set for to first page
ie. page 2 would have an canonical url of "http://www.example.com/category/" and a url of http://www.example.com/category/?page=2

EXAMPLE 3
pagination pages
NOINDEX,FOLLOW
and set Canonical Url to first page

EXAMPLE 4
Category with rel = next and rel = previous links in header
pagination and category page set to "VIEWALL"
ie all pages canonical url of "http://www.example.com/category/?view=all
INDEX,FOLLOW

[edited by: Robert_Charlton at 11:18 pm (utc) on May 24, 2015]
[edit reason] Changed example domain to example.com to disable autolinking [/edit]

[aakk9999]

I normally go for Example 1.

Example 2 and 3 should not be used - this will result in Google potentially ignoring the content of the page (or not getting to these products if you are paginating product listing)

For example 4, if you have viewall page and canonical of other pagination pages point to viewall, then you do not need rel = next and rel = previous.

[dannyboy]

Example #1, but noindex,follow page 2 and beyond.

[seoskunk]

Thanks for the replys Aakk9999 and Dannyboy

OH Dannyboy the pagination, the pagination calling.........

Why noindex,follow?

I would be really grateful if you could explain more.
Its really not clear to me why you would take the time to write prev, next tags, form a unique canonical and then noindex it, I am not worried about duplicating titles , or descriptions so am somewhat puzzled.

[dannyboy]

@seoskunk

Rationale here:
[searchengineland.com...]
[moz.com...]

Just do a text search for "noindex" within those pages to jump to the relevant sections.

Also, this approach is used by the WordPress Yoast SEO plugin: [wordpress.org...]

[EditorialGuy]

Here's what Google has to say:

[support.google.com...]

For what it's worth, I've been using rel="next" and rel="prev" in the manner described Google for nearly two years now, and it's been a "win win": Google sees that the entire series of pages is a single in-depth entity, but individual pages continue to rank well (in some cases,even better than before) for their subtopics.

[seoskunk]

Ermmm
Vannessa Fox (ex Google) on the same site posts differently see [searchengineland.com...]

Thanks for the info Dannyboy

[seoskunk]

Thanke EG good read, maybe this answers the question of whether to noindex or not.

This markup provides a strong hint to Google that you would like us to treat these pages as a logical sequence, thus consolidating their linking properties and usually sending searchers to the first page.

Here's the important bit "and usually sending searchers to the first page", this indicates to me they want the pages to show index,follow so they can direct users to a pagination page when appropriate,

[EditorialGuy]

Here's the important bit "and usually sending searchers to the first page", this indicates to me they want the pages to show index,follow so they can direct users to a pagination page when appropriate,

Also, what would be the point of noindexing "inside pages" when the purpose of pagination is to "treat these pages as a logical sequence"? If you've got a five-page article about household pets that consists of an introduction followed by pages about dogs, cats, fish, and birds, would it make sense to tell Google, "Ignore the actual content of the article and index only the introduction?"

[aakk9999]

Here's the important bit "and usually sending searchers to the first page", this indicates to me they want the pages to show index,follow so they can direct users to a pagination page when appropriate,

You are correct in your thinking. The "noindex" was a solution before the introduction of rel=prev / next attributes.

If you use rel prev/next, do not noindex these pages.

[webcentric]

The above makes sense for paging through an article but what about paging through a blog-roll e.g. first 10 posts (or leads) on page one, next 10 on page 2, etc? The content of each page actually changes over time as new posts are added. Even if the blog roll is a narrow category, it doesn't make as much sense to use rel = next/prev because these posts are not one entry but many. They may be associated logically but they're not a single, static document per se.

[ZydoSEO]

Definitely #1... and I would avoid NOINDEXing page 2 and beyond. By NOINDEXing them you're defeating much of the purpose of rel=prev/next.

The advantage of rel=prev/next is that Google considers the content of all of the paginated pages in the series (as well as the backlinks of all of the pages in the series) when ranking the paginated series.

For example, what if you had a 3 page article. Google will essentially combine the content of the 3 pages into a single page for ranking purposes. So if page 2 has some great content about "subtopic x" that would make the paginated article series rank well for "keyword phrase y", the series might rank for that keyword phrase. In fact, Google might even choose to show the URL for page 2 of the series in the SERPs for "keyword phrase y" rather than showing the URL for the 1st page in the series and then expecting the user to hunt for the page in the series that is relevant to their search phrase. This is why the rel="canonical" HREF on page 2 in the series should point to the clean page 2 URL. By NOINDEXing page 2 and beyond, page 2 will never show in the SERPs.

Similarly, if you have a series of pages that display a particular product (say, product x) from several manufacturers and manufacturer Y's list of product X products appear on page 7 of the series, you want Google to be able to show the URL for page 7 in the SERPs should a user search for "<manufacturer Y> <product x>" type phrases and your paginated series happen to rank well. Your conversion rate "should" be much higher if Google showed page 7 than if they simply showed the page 1 of the series and then expected the user to search through many pages in the series to find that manufacturer on page 7.

[bumpski]

Quoting Google:

Searchers commonly prefer to view a whole article or category on a single page. Therefore, if we think this is what the searcher is looking for, we try to show the View All page in search results.

When Google crawls the New York Times they add something like "page=all" which the Times supports. I see Google crawling sites with these guesses as optional parameters, trying to fetch the entire content.

As a web user I tend to really dislike paginated content, especially sites that paginate down to a paragraph or two or three. The intent is obvious; show more ads, and the unfortunate result is the visitor is very annoyed attempting to return to there original search.

With the advent of mobile, pagination must be torture!

Put a few internal page links at the top of the content to the various sub-topics of your page's content and leave it at that.
If you are writing a novel use pagination!
...

[EditorialGuy]

As a web user I tend to really dislike paginated content, especially sites that paginate down to a paragraph or two or three. The intent is obvious; show more ads, and the unfortunate result is the visitor is very annoyed attempting to return to there original search.

As with most things, it depends. Not everything that's paginated is a "25 [whatever]" slide show. If you were to write a detailed article on breakfast pastries, it might make sense to have an introductory page, a page about doughnuts, a page on muffins, a page about croissants, etc., each with a photo, explanatory text, maybe a recipe, and some links.

That paginated article structure would be much more user-friendly than a page with 2,000 words and a dozen photos that scrolled like a roll of toilet paper.

It also would be friendlier to searchers, who'd be taken directly to a page on crullers or crumpets instead of being sent to a page where they had to scroll, scroll, and scroll again until they found the appropriate section.

[bumpski]

Ah the Anchor Tag or Named Anchors #

Old posts by Google, but still apropos.

"But what if only one section of the page is relevant to your search?"

[googleblog.blogspot.com ]

"Using named anchors to identify sections on your pages "

[googlewebmastercentral.blogspot.com ]

Anchor Tags allow a visitor to easily and randomly navigate large pages with many subtopics. Google will even link to these sub-topics within a page. Once a visitor arrives at one of your intermediate topics, Anchor Tags also allow direct access to the Topical Overview typically at the top of the page. This format gives your visitor almost random access to all your content. There is no large amount of scrolling required! And the sub-topics can actually be labeled

I've provided a Bottom of Page Anchor Tag on some sites and am surprised at how often people use it. Clearly they are looking for the About page or Contact Page, etc (Just like the bottom of this WW page!) Oh and you can throw an ad right in the middle, if you're of that ilk.

Again for Mobile, I would think the Anchor Tag solution would be better than pagination.

Webmaster World supports a strange sort of anchor tag as well:

https://www.webmasterworld.com/google/4748904.htm#msg4748906

https://www.webmasterworld.com/google/4748904.htm#msg4749432

Only you can't enclose them in a post! (and have them function.)

[bumpski]

Thought I'd follow up with a Google search result that when clicked goes directly to a named tag on the page. The page is from the Kaiser Family Foundation, hope that's OK.

https://www.google.com/search?q=social+security+aca [google.com]

Currently the first Google result will go directly to one of many questions asked in this FAQ (KFF is a little slow). I believe KFF used the "id" tag versus the "name" tag in this case. Google's link contains not only the URL of the page but the # (pound sign or hashtag) followed by the "id" tag to vector to on the KFF page. In the future this example might require more keywords, but it works today.

For reference a link to the KFF page itself:
[kff.org...]

The link Google will redirect to for the KFF page with named tag:
http://kff.org/health-reform/faq/health-reform-frequently-asked-questions/question-im-62-and-already-collecting-social-security-are-my-social-security-benefits-counted-in-determining-my-eligibility-for-subsidies-in-the-marketplace [kff.org]

As one can see I'm not a big fan of pagination.

[Selen]

From the link above from support.google.com:

Do nothing. Paginated content is very common, and Google does a good job returning the most relevant results to users, regardless of whether content is divided into multiple pages.

Use rel="next" and rel="prev" links to indicate the relationship between component URLs. This markup provides a strong hint to Google that you would like us to treat these pages as a logical sequence, thus consolidating their linking properties and usually sending searchers to the first page.

IMO the 'Do nothing' option is the best. Sending users to the first page is often not the best idea. It's like the question: "Do I use the meta tag description or let Google write it." In the long term Google will do it better.

Besides, creating next/prev links adds to the code and slows down the page without much benefit.

[EditorialGuy]

Besides, creating next/prev links adds to the code and slows down the page without much benefit.

Slows down the page? I don't know how long your URLs are, but on my pages, the next/prev links add up to several dozen characters at most. (Usually not even that much, since I normally use relative links.)

As for benefits, why would Google support link rel="prev" and link rel="next" if doing so was pointless?

[Selen]

It's not only about the size of the HTML; it's also about running a query to build the prev/next links (which adds to the total loading page time).

Google supports the prev/next as much as it supports the 'Do nothing' approach. Considering the fact that the 'Do nothing' approach was posted as the first choice may hint it's the best approach for most webmasters. I believe soon the prev/next discussion will be similar to the benefit of the meta keyword tag.

[bumpski]

I did find this Q & A answered by a Google staffer, BUT DON'T click on it!
WW removes the "#" from these links (Why?), so you must copy the link text and paste it into the address bar:
https://productforums.google.com/forum/#!topic/webmasters/YbXqwoyooGM

I DID FIND THIS LINK THAT WORKS! (But produces a different format result)
by Maile Ohye 9/15/11
[productforums.google.com...]
Maile makes several posts in this thread. Search for her name. Is it dated info? Don't know!

I do remember pagination, or lack thereof, and Panda mentioned together frequently.

[edited by: Robert_Charlton at 12:06 am (utc) on Jun 2, 2015]
[edit reason] Made first link not clickable. [/edit]