“Quantiphobes” be forewarned. Marketing metrics are about to move to the forefront. The predictive power of advanced statistical analyses used to calculate risk in the credit and insurance industries for years are quickly becoming an integral part of marketers’ jobs.
According to an Association of National Advertisers survey conducted in partnership with Interbrand, 80 percent of CMOs and senior marketers say the board and C-suite are increasingly demanding that marketers be more accountable. And marketers should welcome the change.
CMOs have a higher turnover rate than major league managers. The average tenure of a CMO is a mere 28 months. But C-level marketers can elevate their roles and demonstrate their effectiveness by better understanding and properly managing data streams to apply the power of predictive analytics. Today’s best CMOs should be able to not only quantify Marketing’s Return on Investment (MROI), but predict future outcomes based on changes in a given set of variables.
Traditionally marketing research resources have been primarily dedicated to retrospective exploration: “We spent X on Y promotion and moved Z widgets.” But companies such as Fiat are applying predictive analytics not only to better understand their customers and their marketing effectiveness, but to predict consumer behaviors. “If we dedicate X resources to moving Y metric, we will sell Z widgets.”
Predictive analytics require a new way of thinking about data. Big brand marketers have traditionally treated data sets as separate, discrete pieces of information:
Fiat and other global brands are recognizing the power of viewing these “data assets” not as separate buckets of information, but rather a sea of valuable data to be mined and analyzed in order to uncover causal relationships and predict future outcomes.
While invaluable to marketers, predictive analysis models aren’t a turnkey crystal ball. There are organizational and strategic challenges inherent in adopting predictive analytics. For instance, information tends to live in organizational silos:
However, an effective predictive analytics strategy requires a more holistic view of the management of information assets, including financial performance metrics, metrics regarding the allocation of marketing resources, behavioral and attitudinal consumer data, competitive data and macroeconomic data. The organizational implications of such a model mean the interdepartmental sharing of information and a “helicopter view” of activities, metrics and data assets.
A comprehensive, organization-wide view of information assets is the first step. Once accomplished, the next steps are to mine historical data, determine the most relevant metrics, set baselines and develop a time-series framework for ongoing measurement and evaluation:
Once the framework is in place, CMOs can conduct multilinear regression analysis to uncover relationships, patterns and trends in the data to drive marketing decisions:
Adopting predictive analytics requires collaboration, discipline and patience. But potential rewards are enormous. Determining what’s probable makes predicting the future possible. And what could be more compelling for a responsible, strategic marketer?
strategicNovember 21, 2014
culturalNovember 21, 2014
creativeNovember 21, 2014
economicNovember 21, 2014
© 2014 Davis Brand Capital. All rights reserved.