• SEO-Blog
  • Startseite
  • RSS-Feed, complete Feed
  • About the Blog, Blogcharts
  • Archive: 2003, 2004, 2005, 2006, 2007, 2008, 2009
• Weitere Postings
  • Google's basic-SEO-infos
  • Jerry Yang leaves Yahoo
  • Clickrequest
  • Lycos is closing its doors
  • On PageRank-changes
  • IndexWatch 11/2008
  • Dates, data and statistics
  • Linkbuilding
  • How is Google reacting to the crisis?
  • Fear, uncertainty and doubt
•  

Revisited: Google penalties

It was only a few month ago that I already published a three-part series on the subject of penalties on this blog. Seeing that, with the help of the SISTRIX Toolbox, I am now able to show a few things more vividly and comprehensively than before, I want to take up a part oft that posting again. To be exact, I want to come back to the penalty that I called “Site-depreciation” in the past. This is when the rankings of nearly every keyword that can be found for a site will be set back. This penalty is usually also called the -30, -50 or -100-penalty, to illustrate the changes in position.

For a part of the SISTRIX Toolbox we regularly query the first 10 resultspages for about a quartermillion keywords. Besides such exciting things as tracing the rankingdevelopments or calculating the visibilityindex, we can also use these data to analyze on which resultspage the hits for a domain can actually be found. Besides the possibility for a nifty flash-animation, this also gives us a first impression of how “healthy” the particular domain is. The norm should be a result that is similar to this one belonging to the University in Bonn: 10 resultspages and the ranking-distribution is roughly uniform. What is interesting are deviations from this model. Those who know the clickrate with dependence on position by heart know that it decline extremely quick – arround the thrid page visitors will rather be a coincidence. A pretty example for a perfect distribution is – quite as expected – Wikipedia: the encyclopedia is on the first page for more than 50 percent of all found keywords. To get back to the subject of penalties, here we have the typical distribution for a site that I believe has received a Google-penalty:

Just so I do not offend any small webmaster I used spardeingeld.de, a domain which shows up regularly in my IndexWatch. Either they will already know or they have the chance to learn something here. It is noticeable that nearly no keywords can be found at positions 1-50 – only after that do we see a strong increase up to the tenth page. Therefore, this distribution is a mirror image of the distribution for a page that ranks great and in my eyes, this is a good indication that the domain is subject to a Google-penalty. We get an interesting distribution-variation if the domainholder and -names are strong brands. Here we have neckermann-reisen.de:

The same pattern as before, just that we have an atypical accumulation on the first page. If we now go and take a closer look we notice that these keywords are alltogether keywords which include the brandname (“neckermann reisen”, “neckermann städtereisen” etc.). It seems that Google is excluding brandkeywords from such penlaties – understandable, seeing that generally these are “navigational searches” where the searchee is expecting exactly those results.

[update 11.11.] Gerald made the whole thing come to life at seo.de in an example by means of chronological progression, pretty animation.