Are sweepstakes enthusiasts and low-income tire kickers the only people clicking on ad network banners these days? Underscore Marketing's president explores this idea.

As Matthew Creamer suggested in a recent Ad Age piece, there's a group of less-than-desirable consumers out there on the web who have a high predisposition toward clicking on ad banners.

"…a large percentage of clicking is done by a small segment of less-than-desirable consumers, i.e., the kind of people who spend a lot of time doing online gambling and sweepstakes and who don't boast big incomes. Bad credit is about the only thing missing from this portrait of the heavy clicker as rabble, a portrait, by the way, that answers the question we've all asked ourselves with equal parts wonderment and pity: Who clicks on these things anyway?"

1998 called. It wants its "big reveal" on online advertising back.

Creamer's editorial is just the latest to suggest that banner advertising clicks come from undesirable consumers. If the resultant "get your torches and pitchforks" grumbling in the blogosphere is any indication, this theory might replace the previous version -- that ad clicks come mostly from Romanian click farms.

But seriously, it's something of an open secret that many, if not all, ad networks have cookie files for people who are more likely to click on ad banners, and that they serve campaigns in need of a performance boost to these cookies. (If, for some reason, you didn't know about this, it should explain the behind-the-scenes optimization wizardry that caused a tenfold increase in clicks from your campaign the day after you threatened to cancel a buy running on your favorite network.)

It may also be true that these cookies correspond to people who we refer to around the office (jokingly) as "tire kickers and sweepstakes junkies." I'd argue that when one looks at a pool of hardcore direct response online ad campaigns in aggregate, it would make sense that most of the respondents to those campaigns would fall into that bucket. I'd be surprised if they didn't. When the widest reaching DR campaigns online advertise free screensavers or the chance to win free iPods, is it any wonder what type of consumer will dominate any analysis of click data?

But wait -- I think we're missing the point here. Consider that clickthrough rates are just a measure of initial interest. Then think about the "secret" sweepstakes junkie cookie files in the context of behavioral targeting strategies. Developing a segment of people likely to click on free offers sounds an awful lot like behavioral targeting to me.

This makes me think that perhaps DR advertisers shouldn't be relying on a "spray and pray" online ad strategy, but instead should be building their own files of people likely to be interested in their clients' category, product or service.

I've often said that one day internet advertising should be like a read through your favorite special interest magazine, where the ads are much more like content than advertising because they're so relevant. For instance, over the weekend, I saw an ad on Facebook for a Counting Crows concert and clicked on it. Facebook knows I'm a fan because I've declared my fandom in my profile. As a result, I saw the ad not as a nuisance, but as something valuable and relevant to me.

What if the band started keeping track of the people who interacted with its ads and its website, and used that cookie file to promote album releases, tours and merchandise?

Sounds an awful lot like a behavioral targeting tactic most ad networks call "retargeting," doesn't it?

That's why I think we're missing the point if we start thinking of ad networks as ad sellers who can deliver only undesirable eyeballs. They can deliver the people who are most important to our brands and products if only we took the time to define those audiences succinctly, using behaviors and interests rather than wild guesses.

Tom Hespos is the president of Underscore Marketing and blogs at Hespos.com.