Multiple on-page links to single domain

[MickEdwards]

Hi,

the main focus of my question is to add a bit more information to my client as to where they are going wrong.

Firstly they have a directory on a third party domain with subdirectories containing all their product listings.  All the links need to be no-follow for sure.

However, from a link juice passing perspective if there are per page say 20-30 links to varying URLs on the same domain is it still the same sort of calculation that the link juice is sufficiently weakened that each link has little value, or because the links are to the same domain there is an extra (potential) filter going on?

[Cyrus-Shepard]

I'm going to take the easy way out and side-step the topic of PageRank, which Alan covers well, and raise the issue of relevance and weighting based on reasonable surfer assumptions.

Generally speaking,

1. the first link on a page passes the most relevance, and
2. links from unique domains pass more equity than links from previously linking sites. Meaning each new link from the same domain doesn't pass as much equity as the first link from that domain. It's the law of diminishing returns. http://moz.com/blog/10-illustrations-on-search-engines-valuation-of-links

From a practical point of view, I try to get the highest link on a page. After that, I don't worry too much about additional links to the same domain on that page. But there are a lot of factors that play into this, and your situation may differ.

[AlanMosley]

no don't think so.

Going by the original algorithm this is what would happen. and all testing I have seen since then show page rank flows much the same way still.

When we say 100% that is 100% of the PR it had to pass, as only 85% of PR is passed, http://thatsit.com.au/seo/tutorials/a-simple-explanation-of-pagerank

Imagine the page has only one link, then 100% would pass to that one page, would google weaken it? then why weaken it if passed to 5 pages?

[MickEdwards]

So external  links (pretending these are the only links) on the page

link 1 example.com/spam1.html

link 2 example.com/spam2.html

link 3 example.com/spam3.html

link 4 example.com/spam4.html

link 5 example.com/spam5.html

would not pass 100% link juice surely?  Google would pick this up as unnatural and weaken them?

[AlanMosley]

Imaging a page with 5 links

link1 to domainA

link2 to domainA

link3 to domainA

link4 to domainB

link5 to domainB "no follow"

60% of link juice will go to domainA

20% will go to domainB

20% will be wasted (no-follow)

The link text for link1 and link4 will pass link text relevancy, the other link text will not pass any relevancy.

Its a bit more complicated than this because the first link passed a bit more then the last.

read http://thatsit.com.au/seo/tutorials/a-simple-explanation-of-pagerank