As Paid Blog Networks Collapse ...

[Michael Martinez]

I'm seeing a huge pullback in the link spam community from the large paid blog networks. Some people have sworn openly they are going "white hat", at least for now.

Not only does the collapse of the paid blog network industry level the playing field for many Websites that are not benefitting from large-scale automated link promotion, it also presents an opportunity for experimentation that may lead to some pretty innovative ideas for new linking opportunities.

But recent changes announced by Google have people wondering how much link anchor text may be helping any more. The March link classifier update follows an earlier link classifier update rolled out in February.

I could swear that someone from Google announced (sometime in the last few months) that they would no longer be counting Hotlinks for Google Image Search but I have been unable to find a source for that. I would be most appreciative if someone could point me to a post or comment from a Googler. I'm starting to think I imagined all that.

In any event, people setting up experiments should be careful to keep in mind that Google is also changing how it identifies and evaluates links, so just trying to come up with an "undetectable" blog network is probably not worth the effort. You'll have to take it to a whole new level and think about how to create value for real visitors.

[chaumi]

I'm finding it difficult to second guess how the 'link value' element of the algorithms is going to end up - apart from the seemingly obvious continued value of high value links - but I'm not seeing (yet) how Google can use other 'signals' to lessen the need to rely on link popularity as the strongest one.

From a links perspective, maybe social links gain more prominence? Although they can be manipulated to an extent, maybe the trick will be finding those social links that can't easily be manipulated. Maybe already only certain types of social signals carry any real weight.

Maybe click through rates? Which might increase the need for better crafted page titles and descriptions (have you seen how many of these are actually quite poor?)

Clearly content value and structure, as we've known for some time.

Time on site, multiple page views.

??

[Michael Martinez]

Link popularity was never the strongest signal in Google's search algorithm. It has just been the one most often abused by search marketers.

I suspect that Website or perhaps page-level behavior metrics are having a greater impact on determining which links pass value rather than what sector they are found in.

[chaumi]

Makes sense and very interesting. I've always thought of value in visitor-on-site behaviour from a perspective of how it impacts my site in the eyes of a search engine ie apertaining to their activity on my own pages.

Certainly gives extra understanding as to why stronger sites would pass value...explains how Google can or could keep their momentum on moves away from the seeming blind acceptance for many years that lots of links + lots of votes...and suggests further clues as to which attributes of a site you might want to be looking at when you're analysing link prospects.

[Michael Martinez]

I think the search engines have to rely on correlations more than we do as publishers. Hence, while a lot of people are saying, "I think my bounce rate is really bad" the search engines are probably looking at several signals together such that a "high" bounce rate may be bad in some situations and not-so-bad in others. Even so, if we don't try to second-guess the search engines but rather focus on giving our visitors what they want AND enticing them to want a little bit more, I think most of the marketing falls into place.

You get the links.

You get the visibility.

You get the trust and the authority.

And all of those things contribute to growth in traffic. It's kind of an up-and-down process, in that you see spikes and dips as the weeks roll by, but I feel that is natural and sustainable.