"Data Driven Thinking" is written by members of the media community and contains fresh ideas on the digital revolution in media. 

Today’s column is written by Darren Herman, Chief Digital Media Officer at MDC Partners' The Media Kitchen.

There was a seminal study done a few years ago by comScore, Tacoda and Starcom USA called "Natural Born Clickers." The study found that certain audience segments have a higher propensity to click and engage with online advertising units. A small group, 8% of the Internet user base, accounts for the vast majority of clicks (85%). This is from October 2009, but I have to imagine much does not change.

When you are running audience-targeted campaigns through your trading desk or DSP of choice, what happens when the target you are buying does not have a high propensity to click?  It's surely a wakeup call.

Time to manage expectations, and here are a few pieces of advice:

1.  If you're purely optimizing on clicks, you're missing a lot of the value. Clicks are not everything but they are certainly a nice contribution to an overall engagement score. There are other ways to engage with online display units beyond just clicks, such as the hotly debated view-through metric, qualitative brand impact studies, and in-banner interactivity.  The view-through metric becomes extremely useful when you are running advanced cookie-stack analysis such as with partners like Visual IQ or Adometry. Qualitative brand impact studies help you measure the "soft metrics" which seem to go alongside most audience targeting campaigns (among the usual suspects are GfK and Knowledge Networks).  As for in-banner interactivity, looking at solutions such as Moat to help understand what the user is engaging with in/around the display unit is helpful. The ability to capture all of this data exhaust and use it to help inform your marketing decisions is not as hard as it sounds (or maybe it doesn't sound difficult) and can be deployed with a few phone calls to the vendors mentioned.

2.  Don't forget about placement. Our first party research shows audience is important for a campaign, almost trumping the importance of context. This data comes from one of our clients who ran a research study in 2012 and has a substantial digital media budget. While audience is super important, audience + context is where the home run exists (high income audiences + 300X250 on the Atlantic, for example).  Context matters in this world. Don't forget that, as is easy to do in a land of DSP, RTB, DMP and other technology-driven acronyms.

3.  Buying an audience does not discern "funnel stage." Make sure to manage the expectation of funnel stage with your agency and clients. If you are purely audience buying and trying to drive conversions, there are probably other ways to drive more efficient outcomes by purchasing by funnel stage (higher intent tactics).  You will have to do a lot of explaining on why the media plan might not be working and we all know this takes a considerable amount of time. Match your audience buys to the funnel stage as much as possible -- creating first party audience segments from organic search (using a DMP such as Audience Manager by Adobe) or search engine keyword cohort targeting through a vendor like Yieldbot (disclaimer: I'm an investor).  Using these two examples, you are able to compete in the lower funnel of audience targeting, which is where most audience targeting generally misses.

There is no doubt that audience buying drives outcomes for our clients, but smart tactics are needed when the CTR or engagement rate drops dramatically from historical ad network-based buys. The removal of the "clicker" segment becomes strikingly obvious!

Follow Darren Herman (@dherman76) and AdExchanger.com (@adexchanger.com) on Twitter.