Google recently posted a little blurb about The ABCs of A/B Testing. I hate to be a wet blanket, but for most of us lower volume publishers, that's a pointless exercise with little to gain.

Unless you have very high volume, you simply don't have enough data to draw any useful conclusions from this kind of test.

Don't get completely discouraged: sometimes a small change can produce visible results. For example, when I added ads that mixed in with the text (like the one just above or below this paragraph), my revenues increased enough to notice. However, it wasn't from those ads: it was the column ads that picked up activity.

Read more about Adsense tracking and testing at Webpronews.com.