aristotle - 1:31 pm on Jul 16, 2012 (gmt 0)
Suppose a site has a lot of unique content, excellent user metrics, has acquired hundreds of natural backlinks, and has numerous repeat visitors who spend a lot of time on the site. In addition, suppose that many people search for the site by its name, so that it has a kind of "mini-brand" status.
But also suppose that this same site is over-optimized from an SEO viewpooint. For example, suppose that the webmaster built a lot of artificial backlinks years ago when the site was new, and also used page titles, headers, and internal anchor text to optimize for certain keywords.
The reason I'm wondering about this is because I have two sites with these characteristics which were hit by Penguin on April 25, and lost a significant amount of Google traffic as a result. I'm convinced that the main reason these sites were hit is because I submitted them to hundreds of free directories back in 2006 and 2007. Although these directories were free at the time, some of them later converted to pay-for-listing directories that charge a fee. I used a form filler to make the original submissions, so that the anchor text and descriptions were always the same. I believe that these "unnatural" backlinks caused these two sites to be penguinized.
Most of the pages on these two sites only fell 1-3 positions in the rankings (for their main keywords) when Penquin hit, and they have basically remained at or near these new positions since then. A few pages fell much further initially and have continued to fall since.
My theory is that the pages that only fell a few positions may demonstrate a conflict between Penguin and other parts of the algorithm. In other words, these pages were negatively affected by Penquin, but the positive signals (unique content, excellent user metrics, etc) prevented them from falling more than 1-3 positions. But the pages that fell much further don't have strong enough positive signals to counteract Penquin.
So this is my theory for why some pages on my affected sites fell much further than others when Penguin hit. It reveals a conflict between Penguin and other parts of Google's ranking algorithm. Any comments?