Click-Through Rates are Just Bumps on the Head
Phrenology is a somewhat dated idea whose believers claim to be able to determine character, personality traits, and even criminal intentions on the basis of the shape of the head. They do this by reading bumps (called “fissures”) on the person’s head. While this 19th-century idea may seem a little kooky, Public Relations measurement and marketing measurement often find themselves in a situation analogous to the practice. Back in the day, people thought that you could measure personality based on the shape of a skull, and it was an easy practice. So easy, in fact, that it was not challenged as the main means of measuring personality. Why change something that simple to learn and do?
For example, a Click Through Rate (CTR) is a popular way to measure the success of an online advertising campaign. It’s measured by dividing the number of people who actually click on an ad found on a web page divided by the number of times that the ad was delivered. This means that if one person clicks on an ad that was displayed for 100 people, then there is a CTR of .01 (or 1%).
A CTR is not quite as ridiculous as counting bumps on a noggin, but it still does not measure much. Despite some extensive research using Google and disclosure data from AOL, there is a common belief that CTRs may not measure what companies really want to know. A basic summary of the work is that while CTRs may measure something, it may not be something that is necessarily worth measuring. So, why aren’t we measuring what we would like to know?
My take on this is that people want what is fast and easy, but not necessarily what is the best measurement. People are willing to use easy measurement devices like CTRs because that is what technology drops in their laps. But do these CTRs, AVEs, and other devices let you know if the ad with more blue color resulted in my clicks by women in their mid-40s? I think not. Not using the correct metrics simply leaves the users scratching their heads, wondering why their simple idea did not help them.
Instead of doing what is easy and “been done” already, companies need to first ask themselves what it is that they would really like to measure about their consumers. Then they need to find a way to measure it. There’s a way to measure anything that a company would like to know, but sometimes it is not the easiest way.

July 14th, 2008 at 12:06 pm
[…] can better demonstrate the effects of banner advertising for branding campaigns online (see also Impact Blog). As While CTRs may still offer a useful measure of sucess for some direct response campaigns, […]