MarketingData-Driven MarketingBehavioral Marketing and the Scientific Method

Behavioral Marketing and the Scientific Method

Choosing the right behavioral targeting partner can be tough. Try choosing one based whether it can implement the scientific method to run your program.

I have to admit I was flabbergasted by the breadth and sophistication of the behavioral targeting technologies presented at last month’s OMMA Global-Hollywood conference. Companies are doing amazing things with data and technology. They’re developing algorithms that promise to minimize the number of ads served and maximize the ROI (define) for almost any industry.

When overwhelmed, I turn to core principles for guidance and solace. In all the interviews I’ve conducted for this column, the one that I’ve found agreement on is that ad creative will ultimately determine the success of a behavioral campaign. This is something I can get my head around.

Developing creative is a complex process, but you can apply the scientific method to its development. You investigate, hypothesize, test, and evaluate. Then, like dancing the Macarena, you start over again.

As I pointed out in “A Conversion Professional’s Dream: Behavioral Marketing,” the measurability provided by most behavioral technologies should make it easy to test for ad creative that wins in the marketplace. However, this is only one part of the process.


With databases full of information, a behavioral marketing partner should be well equipped to investigate and target the right prospects. Sophisticated behavioral technologies will match your audience’s behaviors with “lookalikes” — potential prospects surfing the Web who have similar behaviors to your audience’s. They can profile visitors based on myriad categories.

Your behavioral advertising partner will initially ask you who you want to target, but then should provide you insights to help you refine your creative strategies.


The combination of your own marketplace knowledge plus your behavioral partner’s insights will yield demographic, behavioral, and anecdotal data. The challenge now is to hypothesize about which messages and creative will move viewers to take action.

For awareness ads, this isn’t much of a problem. The ad need only draw the eye, leaving the visitor with an impression of the brand and what it stands for.

However, behavioral campaigns are most valuable when they generate action. In this case, ad creative must do more than target. It has to anticipate motivations.

Touch points, or what Richard Rosen, author of “Convergence Marketing” calls “brand interaction points,” help us understand motivations. Knowing where someone saw an ad goes a long way toward telling us why that person is visiting a particular touch point at a particular time. With this knowledge, we will better infer their need for information, action, or entertainment.

For the most part, behavioral targeting doesn’t tell us specifically where our ads will be viewed. Behavioral targeting is about aggregating audiences from a large number of sites. This means our ads could be viewed by surfers with a wide range of motivations.

An extreme example of an ad poorly matched to content can be found on the Fail Blog. Behavioral advertisers are defining segments. Creative must define experiences. If your behavioral partner can target the sites on which your ads will run, your hypotheses will be better, and you will more quickly find the creative that produces a maximum return.


While it doesn’t make sense to test every possibility, behavioral advertising allows us to test a wider variety of creative. A behavioral advertising partner should be able to run a selection of creative through a series of A/B tests and provide specific results. This provides a statistical basis on which to select ad creative.

However, we’re not just testing ad creative. We’re also testing the destinations: the landing pages and microsites that catch prospects. That’s where the action is.

A behavioral advertising partner needs to be up to the challenge of communicating as these tests unfold. A split test may run only until a statistically valid number of clicks have been achieved. However, the ad with the highest number of clicks may not convert well. Which candidate should they declare the winner?

If you want to test your way to success, you must be able to integrate tightly with your partner. It must be staffed to co-manage the tests with you. Look for a partner that can integrate with your analytics systems so it knows quickly which ad converted best.


Once the results are in, it’s important to understand why one ad performed better than another. What motivations does each result uncover? What was it about a particular site that drew more conversions? The answers will influence the next set of ads, landing sites, and strategies to be run through the process.

It’s the ad we liked the least that will inevitably pull the best. Understanding the flaws in our thinking is a valuable exercise that will pay dividends in future campaigns.

Organizations in the behavioral landscape, such as Dynamic Logic, can provide this kind of information. They have the research capability or industry-specific experience to help us get a peek into the mind of the prospect.

If you find it difficult to select a behavioral targeting partner based on data, technology, and inventory, consider choosing one based on how it communicates. Your best choice may be the partner that will help you define your prospects, tell you where your ads will run, integrate with you during tests, and give you some insight into why prospects behave the way they do.

Join us for a one-day Online Marketing Summit in a city near you from May 5, 2009, to July 1, 2009. Choose from one of 11 one-day events designed to help interactive marketers do their jobs more effectively. All sessions are new this year and cover such topics as social media, e-mail marketing, search, and integrated marketing. Register 30 days in advance and get a $40 discount!

Related Articles

Five of the most exciting digital innovations in grocery retail

Data-Driven Marketing Five of the most exciting digital innovations in grocery retail

2y Dev Mehta
How to take your geo-fences to the next level

Data-Driven Marketing How to take your geo-fences to the next level

3y Benjamin Spiegel
Has targeting gone too far? How Amazon ruined Christmas

Data-Driven Marketing Has targeting gone too far? How Amazon ruined Christmas

3y Mike O'Brien
They laughed when I said neuromarketing

Data-Driven Marketing They laughed when I said neuromarketing

3y Dave Lloyd
Listening To Your Customers: A Kind Of Different Intent Data

Data-Driven Marketing Listening To Your Customers: A Kind Of Different Intent Data

3y Som Puangladda
Why the Mathematician Is the New Creative Genius

Data-Driven Marketing Why the Mathematician Is the New Creative Genius

3y Jonathan Attwood
Marketing Buzzword Wars: Personalization Abandoned for Relevancy and Context

Data-Driven Marketing Marketing Buzzword Wars: Personalization Abandoned for Relevancy and Context

3y Jason Klein
Make or Break for Apple Watch Marketing Potential

Asia Make or Break for Apple Watch Marketing Potential

3y Mandeep Grover