The nofollow tag seems to create a bit of confusion among developers, so I thought I would clarify my understanding of the tag and would welcome a discussion.
I have a very easy way of deciding these things for the company I work for – anything that comes through the advertising department should get nofollowed, and anything through the editorial department should be followed. I should point out that I am currently working for a magazine publisher.
The rule of thumb I work to is that if a link has been paid for, the nofollow tag should be applied. Any naturally occuring link within the website content should be left alone.
When the link metric was first introduced as a ranking factor it was seen that a link was a in essence a ‘vote’ for the recipient website. This was easy to manipulate by buying links.
Google combated this by penalising sites that were obviously selling links and followed that through to penalise the websites listed on them.
A natural website linking pattern would should links that are both followed and nofollowed.
There was a thinking that nofollowed links were useless as no ‘link juice’ was passed to the recipient website, but I have always subscribed to the theory that a good link is worth having regardless of the nofollow tag, and it seems I may finally have the proof!
http://www.seomoz.org/blog/ranking-signals-hiding-beneath-the-surface
Randfish, a well known and respected blogger in the SEO field has stated that he believes that nofollowed links from trusted sources will have an impact on your linking profile.
The bottom line is that as long as you are targetting good links, it doesn’t matter if they are nofollowed – a nofollowed link from a trusted website is a good thing to have!
I was asked to help out on a website which has dramatically lost rankings since changing from a non-www domain version to the www version – the same domain, I should stress.
The .htaccess file had been setup correctly but as the dramatic drop in traffic happened 2 days after the redirect happened, it had to be the culprit.
Something that stood out, in the Google listings was that the homepage of the domain was still being listed as the non-www version despite other pages now showing up as the www version.
After some careful analysis of the website it transpired that the CMS used absolute links for all internal linking and all of those links were using the non-www version of the domain.
This essentially would be a confusion for Google as the .htaccess file was telling it to use www links, while the website itself was using non-www – in essence, creating an infinite loop.
With this now fixed I am waiting to see how the big G reacts and if this simple yet essential attention to detail is going to get the traffic back to the site.
The moral of this is that you shouldn’t mess around with domains unless it is really necessary – and you know exactly what is required!
