NEW YORK, May 24, 2012 (GLOBE NEWSWIRE) -- One of the world's leading food manufacturers commissioned a six-figure international market research segmentation study for a top cookie brand from a prominent West Coast research firm; the results lacked bite. One of the most frequently commissioned types of marketing studies is segmentation; smart companies recognize that properly executed segmentation studies increase the impact of marketing efforts by focusing resources more effectively. However, as segmentation studies have become commonplace, their execution has become more rote, resulting in research disaster. In this case, the research firm delivered segment results that just didn't make sense. For example, in one segment, low-income women purchased both inexpensive cookies and super-premium cookies. Decision-makers needed to understand why. "We can't roll this out!" one manager complained. "I wouldn't use it—why would anyone here?"
The company needed better answers—they called in The Modellers, a leading full-service market research firm focused on the application of advanced analytics and predictive modeling to better decision-making—and research disaster recovery.
"We often find that when a segmentation has gone wrong it's because there's a disconnect between statistical analysts who are running models and people responsible for business implementation," indicated Jeff D. Brazell, Ph.D., CEO of The Modellers. "Preliminary interpretation is often left in the hands of junior analysts who are talented in statistical methods but have less business experience. Lacking deep understanding of the client's business, the analyst is left to construct models based on statistical sense only. Then, more senior people are presented with solutions that don't make sense, but they do their best to backwards-engineer a story around the patterns they are given. Often, it's a recipe for research disaster."
As The Modellers dug into the data and segmentation solutions, they recognized that purchase occasion and emotional elements had not been correctly incorporated into the analysis. The Modellers re-ran solutions. The new results made sense—for example, some Moms purchased large packages of inexpensive cookies for their children so they could feel they were "being a good mother," but they also bought a small package of super-premium cookies for themselves, that only came out after the kids were in bed—a private reward. After other segments were fixed, the result was a rich segmentation that was both understood and actionable.
"Successful segmentation doesn't happen using cookie-cutter approaches," Dr. Brazell concluded. "You need business sense infused into every step of the process. Each solution must be evaluated with a business eye to get at validity and actionability, then the solutions have to fit the company where they'll be implemented."
Join The Modellers for a webinar "5 Biggest Segmentation Disasters and How to Avoid Them" on May 30 and June 7. They'll review critical steps every company should take to get better segmentation results.