I decided to write this post because starting from April '08 to today, there have been more and more frequent cases of yo-yoing SERPs.

But it's not the first evidence of Google's yo-yo on a big scale. Here in Italy Michele De Capitani realized and studied Google's yo-yo effect from 2007 and called it "sine waves". In 3 articles Michele explains very well how he discovered this destructive effect by monitoring different Google data centers for different websites.

The initial idea was that previous sine waves and those of March 2007 were the effect of a deep spam check and a method to clean the Google index. But after Google's Phrase Rank introduction and next sine waves of July 2007 this idea was abandoned. Between 2007 and 2008 in Google Groups and other specialized resources in the web you could more frequently read of strange behaviour of web pages: up and down in the ranking with differences, also of many positions, and these weren't a -950 over optimization penalty.

But first of all... how can you realize your site is yo-yoing?The most common effect is a sudden and huge loss of traffic. Yes, you can lost up to 90% of your traffic from one day to the next and your site can still remain in fluctuation for months. Your site will be visible in SERPs only for some keywords and only for short periods (days or hours). Then your site will disappear from high rankings and will be positioned lower and lower until it will be "lost."

Sine waves are related to single keywords and queries. So if you use a good web analytics software or Google Analytics, the simplest method to realize if a keyword is yo-yoing is this: check the keyword traffic for an extended period of time, like one year. If the keyword is fluctuating you'll see large peaks in your graphics. Sometimes after months your site could return in the SERPs, taking its original rankings even if you've done nothing. But many times your site could remain in the waves and continue to fluctuate.

Nowadays it's not clear what is/are the factor(s) that start a yo-yo effect. After analyzing different cases, we realize some common elements, like:

• All kind of sites can fluctuate, such as blogs and websites of different dimensions (from small to big portals)
• Problems of near duplicates
• Problems of network sites in the same machine or in the same subnet network IP addresses
• Problems of websites hosted in countries outside the TLD domain suffix
• Backlinks with the same anchor repeated many times
• Intensive activities of Google's spiders in the site

The latest big fluctuation we observed started on August 19th and its effects are still active.

So... is it the effect of a new Google algo tested in previous years and is now in production at 100%? Or are the fluctuations of the yoyo-effect sine waves the result of a glitch or problem of damping factor? I believe the yo-yo is a new algo completely introduced in production by Google just before the Dewey update....and what do you think about it ?