SEO Your Outgoing Links!

Well said Kev, I agree 100%

Nothing personal Paul, I'm sure you're meaning well, but let's cut to the truth.

This is basically hearsay, and the example of SEOMoz is one based on "belief from authority" instead of evidence. Please link me to their study so I may examine it and scrutinize it in every little detail.

In fact, since this hypothesis has been so commonly circulated as being a fact, I will be conducting the following experiment.

The experiment will be this -

Select a keyword phrase that has a search volume of at least 10,000/month.

Register 6 domain names with private registration, each with unique class C IP addresses but from the same registrar (to avoid any potential trust issues with different registrars). Each domain will have one article on the homepage, with the title tag as the keyword phrase & 400 words of content.

In the middle of the article, 3 of the domain names will have the keyword phrase pointing to a relevant, high authority and high page rank website that is related to the article (let's call these 3 sites class A).

The other 3 domain names will also contain the same keyword phrase in their article, however they won't link anywhere, they will just be left as plain text (let's call these 3 sites class B).

Once this is done, all 6 of the sites will be placed on a page of at least a PR3 ranking. A site from class A will appear first on this page, then a site from class B will appear second, followed by a site from class A, then class B, etc. This way, none of the sites from class A or B will appear any "higher" than the other. There will be unique & relevant content on this page separating each site so it looks natural.

Since all the sites will have the exact same incoming link source, their rankings should be either similar or random; one of the two.

If however, all sites from Class A outrank all sites from Class B under the keyword term, then this will indicate a possibility that an outgoing link to a high authority site may indeed help rankings. If however the rankings of the Class A & B sites seem more randomized, then further studies will need to be conducted, not by myself, but by someone else that believes in this so called "theory".

I'm not setting out to disprove anything, I'm simply seeing whether the hypothesis mentioned by Paul, and so many other IMers, has any merit. If it does, I'll be conducting more research to confirm their findings. If my experiment doesn't indicate anything above chance, I won't bother conducting any further studies, and will instead request that Paul, SEOMoz, and anyone else who believes in this theory provide some evidence to back it up.

I'll be posting my results here hopefully within the next month.

I realize this study isn't ideal. For a start, the sample size is too small, and the linking will not look natural. Nevertheless, it should give us all some vague idea whether this "hypothesis" is most likely true, or whether it will continue to remain unproven. If it still remains unproven, the onus to prove it rests not on me, Chris, or anyone else who questions the hypothesis, but rather on those who propose it.

In the interim, please provide some credible studies that demonstrate outgoing links to high authority sites increase a sites rankings. If not, I will continue with my study and post my findings.