Or maybe “Page Rank Sculpting Now Works Worse than Ever”
Matt Cutt’s latest video clarifies (again) his position on the use of the rel=nofollow attribute within your site. This time, it’s bad. It does harm. To quote Matt, “it does more harm than good“.
Like all SEO consultants out there (sans a very few), I have always advised webmasters to avoid the use of the nofollow attribute. Actually, that’s sarcasm. I kid. It seems these days everyone is claiming to have been an SEO since 1997 and to have never suggested the nofollow attribute was a good idea. Sad, but true.
I have never advised a webmaster to use the nofollow attribute on internal links. I have always stood firm in my belief that using nofollow on internal links was a bad idea, for several reasons.
And I have shared those reasons openly with my clients, whenever appropriate.
With the new “Caffeine” infrastructure, Google is able to annotate links in the web graph better than ever. Matt’s communications suggest that Google is reading the internal nofollow as an important signal, which it is assigning to the link in the link graph. So internally using the nofollow attribute impacts the external view of your links, as far as Google is concerned.
Rational thinkers suggested this years ago… if the nofollow attribute was intended to suggest a link could not be trusted, why tag your own links as un trusted? That’s what Matt is saying now… it will do more harm than good. Perhaps Matt had difficulty clearly communicating this in the past, because Google was not able to cleanly implement it in the past? Don’t know.
Not-so-rational thinkers assumed a level of granularity for the nofollow signal, which may or may not have been actual. They assumed it worked on an instance of a link… such as an internal link on a home page, without necessarily impacting the interpreted character of the destination URL itself (or the host site itself).
I never saw any real data suggesting “page rank sculpting” ever worked, but lots of casual references and, perhaps most importantly, sincere, authoritative recommendations IN FAVOR of its use from high profile SEOs like Rand, Stephan, and Bruce.
I’ve said it before. When you have a vested interest in having been right (because you have clients who paid a lot to be told to implement nofollow, and paid a lot to implement nofollow for page rank sculpting), you tend to develop a biased perspective.
In a cab ride home from SMX Advanced in Seattle, I asked an SEO friend the following question about a high profile SEO consultant who sells an expensive SEO toolset:
“If Google changed something and that expensive toolset ceased to work temporarily, while the SEO tool vender re-writes it or otherwise updates the tool for the New Google, do you think the vendor informs it’s customers that the tool should no longer be used until it is fixed? Do you think the sales department stops selling the tool? Do you think they say
Sorry, we’re not selling our tools right now because they don’t work. But we’re writing new ones! They’ll be ready in a few months…
Of course they don’t. Caveat emptor.