Way back in 2005 Google brought the nofollow attribute to the web community with the sole aim of combating web spam. Hailed as a potential cure for web spam could the nofollow attribute be negatively effecting your web site.
For those not aware, the nofollow attribute was introduced so webmaster had a clearly defined way to tell Google bots and other search engines that a link was un-checked and therefore potential spam or malicious. Links containing nofollow would therefore be ignored by the Google bots, no benefit or penalty could be obtained from a no followed link. The idea behind this was that if no benefit could be given from such links then by applying nofollow to comment links for example this would deter spammers from posting.
The problem was that very soon after its creation the nofollow attribute was adopted by black hat SEO types who would use it to effectively sculpt the page rank of a website. The reason this was possible was that when it was originally created nofollow effectively told Google to ignore all links which contained the attribute. So for example if a page contained 10 links but 9 links were no followed Google would treat the page as if it only contained 1 link. As a result any page rank would be passed on as if there were only one link on that page. Using these methods it was possible to artificially push all your page rank onto specific pages.
Quickly these issues were picked up by Google who immediately changed the way in which it treated nofollow. It is these changes which may be causing you unknown problems.
So What Changed
Basically Google changed the way it treated nofollow links. They still do not pass on page rank to the destination page but they are counted and taken into account when allocating page rank from the source page. So if we take another look at the example above, 10 links of which 9 are nofollow and one is dofollow. Using the new algorithm the page rank is now divided equally between all ten of the links regardless of whether they are dofollow or not. Therefore the dofollow link in this case would receive only 10% of the page rank. The other 90% is passed to the nofollow links which we have already established do not pass on page rank or any other type of benefit. Therefore in this case 90% of the source pages page rank has been lost.
These changes are of little important in terms of external links from your website. The changes do however throw up some fairly major problems for webmasters that have used nofollow for internal links on their sites.
Lets look at an example that I see quite often around the web that is the nofollow attribute being used on login and register links. Presumable this is being done as the webmaster mistakenly thinks this will stop the pages from being indexed and / or the pages will not acquire page rank from the source which we have shown above in not the case. In fact what is actually being achieved here is that page rank is actually leaking from your site. This is happening on every page and on every internal link which is nofollow, that is potentially a lot of juice right there.
My advice don't use nofollow on internal links at all, remember nofollow is not used to block a web page form Google this is done with the robot.txt file or meta tags period. Where possible replace nofollow with good old fashioned moderation, your readers will thank you for it.
Let me know via the comments if you have an opinion about nofollow or think that it does have a time and place within a website.
blog comments powered by Disqus