ECommerce Problem with canonicol , rel next , rel prev

[Saijo.George]

Hi

I was wondering if anyone willing to share your experience on implementing pagination and canonical when it comes to multiple sort options . Lets look at an example

I have a site example.com ( i share the ownership with the rest of the world on that one ) and I sell stuff on the site

• example.com/for-sale/stuff1
• example.com/for-sale/stuff2
• example.com/for-sale/stuff3
• etc

I allow users to sort it by date_added, price, a-z, z-a, umph-value, and so on . So now we have

• example.com/for-sale/stuff1?sortby=date_added
• example.com/for-sale/stuff1?sortby=price
• example.com/for-sale/stuff1?sortby=a-z
• example.com/for-sale/stuff1?sortby=z-a
• example.com/for-sale/stuff1?sortby=umph-value
• etc

example.com/for-sale/stuff1 **has the same result as **example.com/for-sale/stuff1?sortby=date_added ( that is the default sort option )

similarly for stuff2, stuff3 and so on. I cant 301 these because these are relevant for users who come in to buy from the site. I can add a view all page and rel canonical to that but let us assume its not technically possible for the site and there are tens of thousands of items in each of the for-sale pages. So I split it up in to pages of x numbers and let us assume we have 50 pages to sort through.

• example.com/for-sale/stuff1?sortby=date_added&page=2 to ...page=50
• example.com/for-sale/stuff1?sortby=price&page=2 to ...page=50
• example.com/for-sale/stuff1?sortby=a-z&page=2 to ...page=50
• example.com/for-sale/stuff1?sortby=z-a&page=2 to ...page=50
• example.com/for-sale/stuff1?sortby=umph-value&page=2 to ...page=50
• etc

This is where the shit hits the fan. So now if I want to avoid duplicate issue and when it comes to page 30 of stuff1 sorted by date do I add

• rel canonical = example.com/for-sale/stuff1
• rel next = example.com/for-sale/stuff1?sortby=date_added&page=31
• rel prev = example.com/for-sale/stuff1?sortby=date_added&page=29

or

• rel canonical = example.com/for-sale/stuff1?sortby=date_added
• rel next = example.com/for-sale/stuff1?sortby=date_added&page=31
• rel prev = example.com/for-sale/stuff1?sortby=date_added&page=29

or

• rel canonical = example.com/for-sale/stuff1
• rel next = example.com/for-sale/stuff1?page=31
• rel prev = example.com/for-sale/stuff1?page=29

or

• rel canonical = example.com/for-sale/stuff1?page=30
• rel next = example.com/for-sale/stuff1?sortby=date_added&page=31
• rel prev = example.com/for-sale/stuff1?sortby=date_added&page=29

or

• rel canonical = example.com/for-sale/stuff1?page=30
• rel next = example.com/for-sale/stuff1?page=31
• rel prev = example.com/for-sale/stuff1?page=29

None of this feels right to me . I am thinking of using GWT to ask G-bot not to crawl any of the sort parameters ( date_added, price, a-z, z-a, umph-value, and so on )  and use

• rel canonical = example.com/for-sale/stuff1?sortby=date_added&page=30
• rel next = example.com/for-sale/stuff1?sortby=date_added&page=31
• rel prev = example.com/for-sale/stuff1?sortby=date_added&page=29

My doubts about this is that , will the link value that goes in to the pages with parameters be consolidated when I choose to ignore them via URL Parameters in GWT ? what do you guys think ?

[Gijsbert]

First of all: did you check this video? http://www.youtube.com/watch?feature=player_detailpage&v=njn8uXTWiGg

• You can set the ? as exclude from searches in Webmaster Tool

• I would always set rel="canonical" to the main page (category page): .

Check how big sites work with this issue.

[Dr-Pete]

Yeah, it gets ugly fast, and even done "by the book" you're often going to need to monitor your index and make adjustments, I've found. That said, the official Google stance (at least the last I heard) is that you should canonical to the page with no parameters and rel=prev/next to the parameterized versions (your 2nd-to-last example):

• rel canonical = example.com/for-sale/stuff1?page=30
• rel next = example.com/for-sale/stuff1?sortby=date_added&page=31
• rel prev = example.com/for-sale/stuff1?sortby=date_added&page=29

See the bottom of this Google blog post:

http://googlewebmastercentral.blogspot.com/2011/09/pagination-with-relnext-and-relprev.html

The other option would be to use rel=prev/next on the paginated URLs and then dynamically Meta Noindex anything with parameters. Honestly, it really depends on what works, and it can take a while to sort out. Also, keep in mind that Bing doesn't handle rel=prev/next quite the same way as Google.

[Saijo.George]

Thanks for your input.

IMHO...If I exclude ? , then paginated pages like ?page=xx wont be crawled , thus the rel=next prev tags on the page are rendered useless.

[Saijo.George]

Thanks Peter .