Point 1: Small samples have huge variations - make sure there is a large enough sample of impressions and/or clicks before drawing any conclusions. I consider less than 1000 impressions as insufficient to draw conclusions.
Point 2: There are weekly, monthly and annual cycles and seasonal effects. Plus my site is growing. Effects of these may hide other effects or produce misleading impressions - try to compare like with like (e.g. Friday to Friday) or sample over a complete Cycle.
Impressions: The number of times my Ads have been displayed on someones screen. I understand AdWords counts repeat displays (e.g. back button or refresh) separately.
No Impressions? Ad is not showing. Possible reasons: Ad rejected - check for Disapproved Ads; No Active keywords - improve bid(s) and/or add value to the target page (and wait!) Not bidding enough to register i.e. none looks at number 139 on the list! - improve bid or add value; Keywords are not used by target geographic area - go learn the language.
Question 1 What does it mean if I get a very large number of Content impressions? How can I work out what keywords are producing them?
Clicks: Clicks cost me money. Too few impressions and conclusions about clicks are unreliable. CTR is the percentage of Ad Impressions that someone chooses to click on - it must be a function of what my Ad says (Ad copy) and where it appears - either where it is listed in a search or where it appears as content.
No Clicks or poor CTR? Check Ad is as intended. Does the Ad "sell"? Is the ad appropriate for the keywords? What is the surfer looking for form the keyword they are using? Does my Ad offer that?
I expect a much lower CTR from Content impressions compared to Search Impressions.
Corrections and additions please.
Then , try the same with Google, etc.. without the content network and see if it makes you money, so on..
Keep the one, that perform the best...
