Forum Moderators: open

Message Too Old, No Replies

Using rel-canonical for dup content that's *not quite* duplicate

Should rel=canonical tags be used for winnowing options on ecommerce sites?

         

tcbee

7:19 pm on Dec 2, 2009 (gmt 0)

10+ Year Member



I am working as w developers on a new ecommerce site that will provide
lots of winnowing options to sort product thumbnails. Take the widgets page:

www.domain.com/widgets.html

From this page, I can refine results based on price, brand, and type. Ex:

www.domain.com/widgets.html&price=1&brand=2 and so on.

PROBLEM: This will create tons of parameters. That's bad for search engines.

QUESTION: Can the rel=canonical tag be used in cases of severe winnowing? Many of these pages will not be exact duplicates to each other (products displayed will depend on attributes selected), but will be pretty darn near to it...

At any rate:

1. We don't want engines to be including these winnowing parameters in their indices. We just want them to see the main widgets.html page.
2. We also want to preserve link equity as much as possible.

Other preferred workarounds for duplicate content, such as 301 redirects or simply fixing the navigation structure itself, can't apply in this case. Note: we've got a firm handle on the main category structure of the site (no dup issues between the main navigational areas) - the problem is just the sorting and winnowing options.

Thoughts, questions, feedback? I would really appreciate any advice forum readers have to offer.

TheMadScientist

4:40 am on Dec 3, 2009 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member Top Contributors Of The Month



I personally usually use noindex on those type of pages...

There is actually a discussion going on in the Google Forum right now about different 'sorts' on 'results' and what to do with the pages and tedster recommends using robots.txt to disallow the pages in an effort to preserve 'crawl budget' and keep your indexed pages spidered more often, but personally in my specific situations, I think the inbound link weight is more important so I use noindex. I think either is 'correct' depending on your specific situation, site, needs, goals, etc, so my advice is keep the preceding in mind and 'choose your own adventure' depending on how each method would work with your site.

tcbee

5:11 am on Dec 3, 2009 (gmt 0)

10+ Year Member



Thanks, that thread was exactly what I was looking for. I had also been considering using noindex and/or robots.txt, but wanted to see what feedback was on canonical questions first because of link equity / internal PR distribution.

It's interesting that you stated you prefer to use "noindex" because it preserves value from external links, whereas I was originally more concerned about internal link balance. In this case, it doesn't appear that there is a solution to that one. Having 40 links on a page is better than 55 (if the extra 15 go to sort pages), but there's no way to really make all those sorting link options disappear unless.... hmm, maybe you put them in flash or something. Cuz nofollow doesn't work to preserve internal PR anymore (or maybe even JavaScript).

Another option is to use Google Webmaster Tools to consolidate certain parameters, but again, that may still not do much for link equity?