ellipsis flag icon-blogicon-check icon-comments icon-email icon-error icon-facebook icon-follow-comment icon-googleicon-hamburger icon-imedia-blog icon-imediaicon-instagramicon-left-arrow icon-linked-in icon-linked icon-linkedin icon-multi-page-view icon-person icon-print icon-right-arrow icon-save icon-searchicon-share-arrow icon-single-page-view icon-tag icon-twitter icon-unfollow icon-upload icon-valid icon-video-play icon-views icon-website icon-youtubelogo-imedia-white logo-imedia logo-mediaWhite review-star thumbs_down thumbs_up

How data can bring creative and media back together

How data can bring creative and media back together Michael Lowenstern

Not long ago, I was in a car with a colleague from a media agency, and we were arguing the merits of media targeting and optimization versus creative targeting and optimization. His claim essentially went like this: "Mike, it doesn't really matter what your ad says, if you give it to me, I'll find the right audience for it and then optimize the hell out of the media."  

I've been in the advertising business for 20 years, and I know the disconnect and distrust between media and creative run deep. So, while this comment wasn't surprising -- I've heard it fairly often -- it still feels shortsighted to me. Advertising involves both art and science; therefore considering just one-half of that equation returns a fraction of the potential. Imagine the results we could deliver if both media and creative were optimized in concert.

One of the benefits (or curses) of working in digital display advertising is the fact that we straddle the two disciplines and must operate fluently within both. This article is my attempt to bring the two sides closer together. I believe the exercise of using data as a catalyst to synthesize media and creative will benefit all parties involved, it will reinvigorate the medium of display advertising, and, yes, it will improve results for our clients.  

Stay informed. For more insights into connecting with lucrative online audiences, attend the iMedia iMoms Summit, May 1-4. Request an invitation.

First, a disclaimer: I'm a creative/tech guy, so to media teams, I will channel Howard Gossage and say, "No amount of targeting and optimization will make an ad interesting -- creative still has a huge impact on viewers." (Study after study has shown that clever, creative advertising improves persuasion measures, such as recall, brand attitude, and purchase intent.) And to creative teams, I advise that denying data is like denying global warming. It's our reality, like it or not.  

In the good old days of broadcast television, one ad was delivered to everyone, regardless of whether they were the right target for the product being marketed -- and it worked. As this medium evolved, networks began dicing up their audiences by demographic, offering more relevant (and responsive) targets to marketers. But, even today, those targets are none too specific, so the advertising has remained demographically hit-or-miss as well; everyone still gets the same ads on TV. For example, this weekend my wife, 11-year-old daughter, and I were offered the opportunity to learn about Viagra, Budweiser, and pickup trucks during an evening in front of the tube. 

The digital space has followed a similar trajectory. Since the 1990s, we have created essentially the same experience. Though during the intervening 15 years vast amounts of audience data were brought to bear, and we have become able to display ads to narrower and more focused bands of viewers consuming the same content, we have still delivered one-size-fits-most creative executions. Despite that, data-driven placements have improved response and efficiency, so media got the credit for the recent boom cycle in display advertising. Those of us on the creative side of the fence worked on our screenplays and acted all emo about it.  

Part of the reason creative teams have been left out of the display advertising strategic loop is that we are naturally resistant to data, analysis, and critique. Another part is that it is very difficult to create template advertising that's interesting and engaging. Yet another part is simply that data can be very prescriptive and restricting. I propose a truce using my favorite Einstein quote: "Not everything that counts can be measured. Not everything that can be measured counts." Translation: Those on the creative side should use the data that makes for interesting and relevant advertising. Like all creative enterprises, ours is a reductive process, but we must include past performance, audience analysis, and predictive targeting in our tool set. We want to connect with people, and these are powerful tools. 

I advise the media side that the data marketplace is fairly saturated now. Everyone is bidding on the same audiences, and we are starting to see a plateau in both media performance and efficiency -- even with behavioral data. At the same time, creative display technology is mature -- we have tools, processes, and mechanisms to develop adaptive, dynamic, and multidimensional ad units. I believe that the next display cycle, starting now, will be about the merger of data, media, and creative. Given the ability to know who is looking at our ad, where the ad is placed, what content that page contains, and its sentiment, our technology will deliver the most appropriate creative version of an ad to those audience members targeted by ad exchanges. It optimizes the creative based on past performance -- and given enough social lubricant between creative and media teams, it could just as easily trigger optimization routines in the media as well.  

So, what's the process? I propose the following schema:

When the campaign is initiated, determine appropriate KPIs.

Every vertical, client, campaign, medium, format, and creative execution has a different set of benchmarks by which to judge its success. Are we delivering a user to a landing page? The KPI could be a "click." Are we calculating CPO or CPA? The KPI might be form completion or order completion. If the goal is to influence purchase intent, the KPI to consider might be interaction rate or brand time. Determining these KPIs allows both media and creative teams to calibrate to the client goals, and develop the most appropriate plan of attack.

Creative and media plan together and agree on appropriate initial "best guess" segments, developing each strategy in concert.

In my experience, creative and media look at target insights quite differently. Given the opportunity to develop a response without being constrained to a pre-approved media plan, creative strategists are more able to find unique and varied message opportunities. Of course, media will bring their recommended target set to the table, though the two overlap to a good extent. By distilling them into the best, most efficient target-set, we can provide the client with the best of both worlds -- and with group ownership of the total plan.

Form a communication matrix based on the creative/media strategy.

The creative teams can use this to develop multiple executions under their conceptual umbrella. This matrix is probably foreign to all but the most direct mail-experienced creative teams. More importantly, it is also foreign to some more traditionally minded clients that are used to approving only one target for any given campaign. So, present these clients with the matrix and the logic behind it -- this will keep the program on the rails as the campaign progresses.

Develop and program technical frameworks that allow for flexible content within each ad.

The ability to create flexible creative "shells" -- or the ability to create multiple versions of standard Flash ads at scale -- is crucial to being able to deliver customized creative without breaking the bank. This part is easily managed with just about any rich media vendor. But without a creative solution that allows a certain level of flexibility, the ad creative nearly always suffers from "template-itis."

Deploy the ads, and after a delivery threshold has been met, begin regularly scanning the DSP audience reports.

Reports that show how creative versions are over or under-indexing (broken out by audience) provide valuable insight not only into how a creative version is performing against our presumed target, but also what new targets may exist. If an ad version is successful with a target that hadn't originally been anticipated, place and weight that creative with this new audience segment. Conversely, if a message version is underperforming, tweak it if possible, or remove it from rotation.

Rinse, repeat.
Again, studies have shown that the combination of relevant targeting and relevant content is symbiotic. If we are to meet the ever-increasing challenge from our clients to improve efficiency and ROI, we must be willing and able to work together to systematically develop and deploy interesting, engaging, and effective display for all of our potential audiences, dynamically and at scale.  

Michael Lowenstern is managing director of digital advertising for R/GA.

On Twitter? Follow iMedia at @iMediaTweet.



Michael Lowenstern’s 15 years as a professional bass clarinetist specializing in new music, and his 15 years as a professional marketer specializing in online banner advertising actually have two things in common: They are both very niche...

View full biography


to leave comments.