|
VPC is nowhere near as reasonable as it may sound. For example, unless everyone is deluding themselves, the GEO ads produced (fairly) recently by AFF were responsible for a big hike in visits to the sponsor. But as a general rule, although an increase in CTR may lead to more sales in total, it usually (although not always) results in a lower conversion ratio.
Earnings per impression is superficially a more appealing yardstick, but that starts getting complicated when you recognize all advertising spots are not created equal. Physical location (which of your sites, which page on a site, which spot on a page) is not the only fly in the ointment: a promo placed near one which is very successful is going to suffer simply by that proximity. And how do you (fairly) compare the results you should expect from a half-page ad against those from a text link placed in a table with a dozen others?
All of which forces me into a more pragmatic approach. I look at the sponsors performance each month, trying to identify those with low exposure that have done better than expected so have earned a higher profile, and those which have been disappointing. I look at the weak promos to see if I can do something to make them work better. The weak ones I don't believe I can fix get dropped and new sponsors or sites get exposed initially according to my instincts.
I don't pretend that is very scientific. On the other hand if someone running his own business doesn't have a good instinct for core details of that business, he/she isn't going to be successful for very long.
|