Pieter Sanders and Bram Lebo wrote an article at Brandweek.com recently talking about the value (or lack thereof) in tracking clicks. They actually performed a case study with (presumably) a real client, and compared the click through rate (CTR) of a given advertising campaign with its overall reach. The answer: they're unrelated at best, and at worst a high CTR typically is associated with a shorter reach. As they found:
"The search engine [campaign] delivered a grand total of 4,000 impressions [with a high click through rate]; though many of those bought the product, in the end only four thousand were exposed to the campaign. Compare that to a million impressions delivered on a content site, albeit at a lower CTR."
This is a very interesting perspective. If you're selling a product, do you only want to expose it to people looking to buy that type of product *at that moment*? If so, then this is the way to go. But what about the guy that's going to be in the market for your product next month? Don't you want to advertise to him NOW, and begin building mindshare? Or would you rather his first exposure to the product be during his search when buying? If he has 4 competing products lined up in his web browser, does it not stand to reason that he'll purchase the one with which he's most familiar?
Lose sight of branding, and you lose that customer.