Secretly back-linking from whitelabel product

[seoczar]

Lets say a company (provider.com) offers a whitelabel solution which enables each client to have all of the content on their own domain (product.client.com), with no branding by the content provider.

Now lets say that client.com is a site with a lot of authority, and to promote the launch of product.client.com, they put a lot of links from their main site to the subdomain. This can be very valuable link juice, and provider.com would like to be able to take advantage. The problem is, that client.com wouldn't like it if provider.com put in links on their whitelabel site.

Suppose the following:

All pages on product.client.com start to have a rel="canonical" link to themselves, with a get variable (e.g. product.client.com/page.htm -> product.client.com/page.html?show_extra_link=true)

When the page is visited with the extra get parameter "show_extra_link" a link appears in the footer that points to provider.com

My question is, would this have the same effect for provider.com as placing a link on the non-canonical version of the pages on the whitelabel site would?

[AlanMosley]

With prev next the content of every page is given to page 1, in that case the link would be part of the content. But with a canonical I am not sure.

If you go by comments by Matt Cutts and Bings Duane Forrester canonicals are the same as a 301 execpt they dod not pyhsiclly move the viewer to the canonical page. so in the case of a canonical the content would not be merged, only the content on the canonical page would be indexed, the links from other verrsions would be redirected. so the link on the show_extra_link version of the page would not be indexed.

As for the morality of this, i would not do it, you are not being honet with the clint and you would be caught out sooner or later when the url was seen in the index(if it was indexed)

[seoczar]

Hi Alan,

All of the pages on the subdomain have a robots meta with noindex, follow on them. The pages are only used for data collection (forms), and the clients do not want their pages showing up in google, which is why extracting link juice shouldn't be a problem. As such, the canonical url need not be indexed.

From what I understand, if a page has duplicate content and specifies a rel=canonical, url, the inbound link juice effectively gets syphoned into the original content page. What I'm wondering is, which page does google use for the purpose of propagating outbound link juice?

[AlanMosley]

i would say the canonical.

if the pages are not indexed, but follow, then they would have no value themselfs unless they had in-coming links. if they do have in-coming links then yes they will pass link juice, but only from the canonical i would think, based one what i said above about a canonical being much like a 301

[Dr-Pete]

I'm with Alan - in theory, the canonical would pass the link-juice to the version with the link, but you're not only misleading the client - you're one step away from cloaking the link. You could actually get your own clients penalized for this, and that seems very short-sighted.

Add the NOINDEX on top of this, and I'd be willing to bet that the value of these links would be very low. Even if the client approved followed white-label pages with footer links, for example, we're seeing those types of links get devalued - they're just too easy to get. Now, you add these links all at once, NOINDEX the page, and canonical to a weird variant, and you've painted a very suspicious picture for Google. It might work for a while, but you're taking a significant risk for potentially a very small gain.