I'm of the impression that the average revenue per click is a false metric created by google to throw off both financial analysts and ultimately the governments in ultimately coming down on google's monopoly on search. It gives them an argument against anyone coming down on their enormous monopoly on search.
Every quarter, quarter after quarter, year after year their ad revenues go up. It costs them bubkis for every new click. Once set up it doesn't matter. It could be a new click on a new ad in a new category that makes them $0.06 or it could be a click on an ad by a personal injury attorney competing for the most aggressive terms at $hundreds of dollars per click. Each time a new ad campaign starts it pushes up revenues and once going it has virtually no impact on costs.
We've been running local smb adwords campaigns since May 2005. I just looked at the longest running campaign. I looked at two telling phrases. One is exact match. It only shows when the exact term is used. Its been running in the same geographical region for over 8 years. Our average cost per click on that phrase is 3x what it was in 2005.
Then I looked at a broad term phrase. That means if that exact term shows or if terms "google deems" similar" for various searches (again its all local). In the first year that "term" or terms close to it generated 1300 impressions. In the last year (12 month period) it generated 47,000 impressions. In the first 12 months that broad phrase match generated 150 paid clicks. In the most recent 12 months it generated about 540 clicks.
From 1300 impressions to 47,000 impressions. WTF. What occurred? Google never told us. That is one thing we all know!!!!
Every time the use of that phrase generates an impression it impacts the other ads in that list. Ultimately it pushes up the cost of the ads. its a bid system. Adding more advertisers pushes up cost per click...and google is adding advertisers galore.
I believe it would be far more telling if google generated a stat that everyone could see as to total impressions shown. That would be a metric more in line with print metrics on total ad pages. It would be incredibly more telling.
We don't get that stat. Nobody gets it. Nobody knows how many times google shows ads. But google knows.