No Follow Links & PageRank Sculpting

Thursday, 18 March 2010

Matt Cutts, who works for the Google Search Quality group and specializes in search engine optimisation, recently published a video where he speaks about PageRank sculpting. He spoke specifically about the practice of annotating hyperlinks to non-essential pages, such as a privacy policy, with the rel=nofollow attribute. Many people subscribe to the belief that this positively influences the PageRank of essential pages.

When Matt Cutts said that it was NOT an influence, it shattered many people’s PageRank sculpting theories and practices. Therefore, this was an assertion that required testing and the tests did indeed show that the addition of the nofollow tag did not affect the PageRank of other subpages. However, it was unclear if this was because it had no effect at all or because the effect was not large enough to increase the PageRank to the next highest whole number, e.g. 4.75.

Another observation made during testing was that 10 hyperlinks linking from the home page would reduce the subpage PageRank by one, e.g. PR5 to PR4. If there were 100 hyperlinks on the home page, using the nofollow tag reduced subpage PageRank from PR5 to PR2 or PR3. This contradicted the common interpretation of what Matt Cutts had said.

It was clear that blocking a hyperlink does have some affect on the PageRank value of other subpages. Moreover, this makes sense because otherwise we could infer that an unlimited number of hyperlinks would produce the same PageRank value. This is of course untrue. It would be interesting to hear Matt Cutts explain this matter in more detail.

Furthermore, testing resulted in the conclusion the hyperlinks tagged nofollow remain discoverable. Google follows these “nofollow” links but does not factor them in to the PageRank value. So rather than “no follow”, it would be more appropriate to call them “no vote”.

In addition, it is important to keep in mind that there are countless other factors involved, many of them impossible to quantify from our end. In addition to PageRank, Google uses on-site positioning, navigation depth, and off-site elements, such as link and contextual references. This makes the task of testing PageRank all the more difficult.

In conclusion, the testing led us to the belief that a webmaster can better spend their time generating content or working out a link deal. Constructing a nofollow structure may not be a waste of time but it is clearly not as valuable.

This post is provided by Dejan from Dejan SEO, a search engine optimisation company specialising in advanced link building techniques.


Carpet Protective Film said...

This is the first I visited your blog and i love it. Bookmarked it already. Keep up the good work.