Hey, Check Out the Heavy on that Clicker.
According to a comScore study commissioned by Starcom and TACODA, online ad clicks aren't as demographically diverse as your deluded CEO thinks.
80 percent of them come from only 16 percent of online users. They are generally young, underpaid and male. You know, like the dev dork of yore.
The people that rack up 80 percent of your clicks are divided into two groups:
o Heavy clickers (which click on ads four or more times a month) constitute six percent of the online population and make up 50 percent of ad clicks.
o Moderate clickers (which click on ads two or three times a month) are 10 percent of the online population and make up 30 percent of ad clicks.
According to ClickZ, heavy clickers are almost evenly divided between men and women between ages 25 and 44. They generally make below $40,000 and are wont to visit auctions, gambling and career service sites. They spend five times more time online than non-ad-clickers.
Looks like behavioral targeting can't become the norm soon enough. TACODA must be creaming itself.
Comments
This article made the rounds at my office last week, and I'll say here what I said there:
While the 6% of users generate 50% of clicks data point is interesting, there are two issues with the findings:
1) The whole CTR conversation is weak. Most savvy marketers abandoned CTR as the sole success metric a long time ago...in favor of a combination of brand/direct response/engagement metrics, depending on campaign objectives. I wouldn't expect there to be a correlation between clicks and brand metrics, as a click is representative of a direct response metric, not really a "brand" one.
2) They fail to provide context for the study: did it include ALL types of display ads (static gif, animated gif, Flash, rich media/expandable/fly out/video/audio)? What kind of message/offers were included? Did those marketers have other media in market? We know that media working together is typically more impactful than one channel on its own. Was behavioral targeting used in any/all instances (as one would assume, since Tacoda backed the study)?