We’ve been running all sorts of tests and experiments with our partner network’s blogs.  It’s only been one week into the current round of tweaking and testing and its already yielding some interesting nuggets of information.  Some findings:

1) Google adsense seems to go on a “freeze” mode if you update any element of the blog that impacts adsense code–themes, sidebars, ad code color, etc

2) Reducing the number of ads in an ad unit doesn’t necessarily boost CPC.  While CTR might increase (depending on your layout), the overall CPC can actually decline… by a lot.  This runs counter to the “conventional wisdom” that CPC rates will go up if you have less ad units since there’s less lower bidding ads appearing–those that appear would only be the winning high bids.  There is a possibility that the prevailing ‘wisdom’ might still be accurate even in the face of our own findings.  How?  The existence of some sort of Adsense “sandboxing” system for system changes.

3)  There’s a correlation between content quality and Adsense CPC.  Some may consider this a nobrainer since the right well-crafted text should be able to pull the right high value adsense ad.  It goes deeper than mere CPC keyword sniping though–the overall value of the content and its appeal to the viewer might be factored by Adsense regarding which ads to show.  We studied sites with high “stickiness” and multipage view or ‘deep browsing’ traffic patterns and compared their Adsense performance with high bounce rate shallow sites.  While not consistent across the board, we are convinced that the difference is substantial enough to warrant further investigation.   Content may be “king” when it comes to traffic…  It appears to also be very important when pulling juicy fat clicks from Adsense.

NOTE:  The partner network that has authorized us to do Adsense optimization and testing has over 100 blogs.  Not all the behaviors discussed above occurred in the same way or in the same rate.  This may be due to the fact that Adsense has differing data centers.  However, there’s enough of a common pattern that we can’t ignore these findings as quirks or one off events.  Anyone SERIOUS about boosting their adsense income would definitely find these patterns and experiments worth further investigation.

Feel free to share your own Adsense experiments’ results in the comments section.

Photo Credits: LUXAMART