A Lot is written about links, linkbuilding, and the general feeling that webmaster have that you have to get as many links as you can to rank well.
The problem with this approach is that if everybody does it, the whole web has to turn into a spamfest of linkfarms with everybody trying to get their sites to rank by generating links as fast as possible, and makes life very difficult to any Search Engine trying to assess a website’s genuine popularity by the amount of links pointing at that site. With an estimated 108 million+ websites (Feb 2007) out there now, if each needs a couple of thousand links to make it rank, you get an idea of the kinds of numbers of Links that Google have to sift through for the real value links.
Hence it obviously makes sense to filter them in a myriad of ways, to either enhance or devalue them as far as rankings go. The Pagerank system did an OK job of this in itself to start with because a higher PR link is simply more powerful due to passing (we’ll call it “linkpower”) or maybe think of it as like putting a higher voltage on the line, it will pass more current down it.
obviously however, as with anything man-made this sytem could be abused so this is where the myriad of complicated and secretive filters are then applied to increase or more likely decrease a link’s strength depending on all sorts of factors we can only guess at.
Google’s official line is that of a company looking to make a better web for the future, relevant, themed and contextual linking is the way forward. We should stop worrying about PR and build the best links for users that we can.
Well, we”ve been in the lucky position of being able to arrange couple of extremely high quality links to a new client site, from one of their sister companies and watch the effects of that alone.
By high quality links we mean from a page with good PR to outbound link ratio, and with our idea choice of perfect anchor text, inserted into the middle of a very relevant body of text mid-page. Really, apart from higher PRs still they dont get much better than this.
We had already made initial changes to the site and had them cached, and let the pages settle to their new natural level. The company were struggling at between #7 and #11 for their own name, as the name is an extremely common term.
The first link came live (cached) made no difference at all for about 3 days, and then over the course of the week, has slowly but surely pushed the site up to a strong looking number One for it’s own name, which was the anchor text of the link.
Now this link has reasonable strength (as in PR/outbound links) but the strength of the SERP rise in a reasonable sized search (7 million pages) in just 2 weeks suggests that this link has maybe even been +ve weighted, which implies that what Google say about relevance does indeed play a big part in link weighting and hence ranking for specific terms.
However, what is also interesting us greatly is that once we have got, borrowing from the excellent Hobo’s terminology from SEO Heat Theory, a “hot” page, is that anchor text that is then assigned to the page from far weaker sources of links, PR and relevance-wise, will also see the page start climbing for those terms in the anchor text, much quicker than normal.
It would definitely seem to us that if you “hotten” a page or a site up with powerful high quality links, that carry no anchor value, that you can then assign a little anchor text to that page from a different place, and being as it’s already “hot” it’ll rank for that too.
We are also seeing this currently wth our SEO Experiments & Theories page where it is (often intermittently) showing first page SEO+Experiments, and second page for SEO+Theories.
There is a very little anchor text directed at this page from a couple of forums, with seo experiments, but none whatsoever with se o theories (broken so it doesnt get auto-linked out from our snazzy plug-in;) )
But being as we have reasonably strong (PR/OBL) links pointed at it (with irrelevant anchors to the seo experiments search) and because it’s a “hot” page, there it is, first 2 pages for the two variations of terms in the pagetitle.
And, on the Spanish Google index, this page is so “hot” it sometimes now often outranks (and replaces) our homepage for ” Search Engine Optimisation in Ibiza ” and is not nearly as relevant. The terms dont actually appear on the page, except as onpage menu anchor text to the homepage.
So all in all, it would appear that our tests are confirming that the value of a highly relevant and powerful link cannot be underestimated, and we would go as far as to speculate that maybe 3- 5 really good links like the ones above would rank your site far better than maybe 2000 spammy links from directories, forums and what have you.
And doesnt it make SEO easier? Google’s vision of the web is much less work for us and far more satisfying to know just that one link you researched, sourced and arranged did the job. And it makes for a nicer web for everybody at the same time.
We applaud them and are embracing the new way forwards, 5 links per client,
sorted 
