Google has recently released its new version of the Google Panda algorithm and they claim it will reduce the amount of spam on the web from tarnishing Google’s search results. Searchmetrics have done an analysis of who has benefited and who has taken a hit from the new algorithm. Some of the winners include marvel.com, epinions.com and menshealth.com and some of the losers include digg.com, appbrain.com and newsbusters.org. It is likely that these winners and losers are showing the effects of an earlier Panda update rather than the new algorithm change.
Matthew Cutts stated:
“Hey Danny, there’s a pretty big flaw with this “winner/loser” data. Searchmetrics says that they’re comparing by looking at rankings from a week ago. We rolled out a Panda data refresh several days ago. Because of the one week window, the Searchmetrics data include not only drops because of the webspam algorithm update but also Panda-related drops. In fact, when our engineers looked at Searchmetrics’ list of 50 sites that dropped, we only saw 2-3 sites that were affected in any way by the webspam algorithm update. I wouldn’t take the Searchmetrics list as indicative of the sites that were affected by the webspam algorithm update.”
Searchmetrics summarised the losers as having some of the following characteristics: They use databases to aggregate information, many of them are press portals and aggregators and they have Heavily-templated websites.
Many people were very shocked to see some of the sites that took a hit. Digg, Techdirt, Paidcontent and Newsbusters are all big names, however, the most shocking site to take a hit was Cult of Mac. The site has previously complained about being hit and then they returned, arising confusion about whether it was manually restored, Google said that it was dropped for other reasons.
Occurrences like this remind us that analyses like these are not always perfect. Sites can have gains and drops for different reasons, they may be less visible this week because the previous week an event or a certain degree of media exposure happened, for example. The parked domain classifier that Google imposed a few days before Panda 3.5 is another example of an inconsistency that can affect the results.
The only way to be certain that your search results have been affected is to look at your search-driven traffic from Google itself rather than listed websites from an analysis that may be inconsistent.