• by Jeffrey D. Mills , November 17, 2009

We have more data on consumers today than ever, but do we really know more about how to market to them than we did 25 years ago? It's not our fault; it just used to be simpler back when the average consumer was easily defined, shopped in fewer places, and print, TV, radio and in-store were the only real options available for communicating with them.

Everyone knows that the Internet has changed the media landscape but we are only now beginning to realize how and to what extent. This past April, the GfK Group, Google & Coca-Cola announced that they'd measured a 97% higher purchase rate when TV and YouTube video interplay were part of an orchestrated, cross-media campaign.

The web is more than just another communications channel. It amplifies your entire marketing communications strategy, accelerates the pace at which consumers adopt and increasingly exposes people to influential messaging that you don't control. We can continue to invest increasing amounts into building bigger, more integrated databases or we can start getting smarter about understanding why consumers behave the way they do.

September 2008's ARF Journal of Advertising Research talks about the rising importance of "Long Tail." The article articulates how companies are increasingly investing in measuring consumer activity in each and every possible niche. I'm not even sure this is possible, but I am sure that it isn't scalable. It entails committing to a never-ending game of catch up and the near-impossible task of staying current. It seems every day there is a new social network, TV show or magazine popping up.

The "single source" approach has been tried, most notably with Project Apollo, which was shut down in 2008 "because not enough clients signed on to make it a viable business proposition going forward." This hasn't stopped companies from coming out with media exposure RFID devices that you carry on your person, biometric devices that "measure" your response to Ads and cell phones that capture the radio stations and movies you want and ones that can send a signal when you've driven by a billboard. This all translates to us as yet another mile marker passed in a race without a finish line.

We need to stop the insane pursuit of all data on what consumers did and shift our attention instead to understanding why consumers do what they do. The reality is that seemingly complex consumer behavior can be sufficiently understood through the collective impact of a parsimonious set of rules. What makes a "rules-driven" approach scalable and sustainable is that you don't need many to explain why consumers behave the way that they do.

The problem with taking historical outcomes at face value is that they ignore the convergence of many things happening that drove those outcomes and assumes the past will equal the future. Looking at the marketplace from a rules point of view lets us look at the marketplace from the perspective of the individual consumer and then gauge the collective impact of marketing at the consumer level (and not just broad aggregations of the marketplace). Most marketing mix models today are built using historical outcomes and, because they're only developed every 12 to 18 months, businesses wind up basing marketing decisions on static information that can be up to two years old. We can no longer rely solely on static, aggregated snapshots of a marketplace that isn't static.

There are 3 significant advantages to taking a rules-based, simulation approach to marketing planning:

1. You actually get consumer insight. Simulating the likely sales impact that would occur when different marketing plans are put to work against a virtual population of consumers reveals how different types of marketing impacts different groups of consumers. Mix models aren't designed to make you smarter about your consumers. They exist to provide you with a post mortem on the aggregated, market-level effectiveness of last year's plan.

2. You get a collective, all-encompassing view of your marketing impact. Knowing that you've increased web site traffic or increased your reach for a TV ad is good, but being able to test how different spend allocations across print, TV, word-of-mouth, radio, digital, promotions, out?of home, etc. work in unison to impact sales transactions is even better.

3. You can quickly run an unlimited number of forward-looking, "what if" scenarios. When you have a model built on the rules that drive human behavior but calibrated to your category and brand, you have a platform that can tell you what would happen if you made changes to your marketing plan and/or if changes occur in marketplace dynamics. Traditional mix models don't provide this flexibility and leave marketers to essentially guess how new variables will impact their marketing efforts.

Businesses can't continue to collect more data in an attempt to better understand a consumer marketplace that is clearly outpacing our ability to keep up. The technology exists today to employ a more responsive and scalable approach to marketing planning that helps marketers understand the collective impact of marketing on consumers. It's time to step back from siloed metrics and get to know the consumer all over again. This time, from the perspective of why they behave the way they do instead of relying entirely on data that tell us only what they did.