Why marketers should target individuals, not households

Alex Verkhivker | Jun 08, 2018

Sections Marketing

Most marketers treat a US household as a single unit for targeted price promotions. For example, a supermarket may offer a shopper, based on data collected about the household, coupons for paper towels or trash bags.

But inferences that marketers make from aggregated purchases across a household don’t accurately reflect the behavior of its individual members, and marketers would be better off looking at the individuals within households, according to research from University of Cologne’s Hernán A. Bruno, Public University of Navarra’s Javier Cebollada, and Chicago Booth’s Pradeep K. Chintagunta. “Targeting policies based on household-level parameters can lead to sub-optimal marketing decisions,” they write.

Understanding how each member of a household behaves is possible today because of how companies now track consumers’ purchases. With the rise of digital platforms such as Facebook and Amazon, companies can both track purchases and link them to other individuals, either from a family or a social network. 

In their study, the researchers used data from a major European retailer, focusing on purchases made by almost 20,000 households during a six-month period, from May to November 2007, in five categories of products—coffee, pudding, bleach, dishwashing liquid, and paper towels. Because a small set of brands account for the bulk of sales in these categories, the researchers were able to home in on factors such as brand loyalty as well as the effect of price changes and targeted promotions on household members’ purchase behavior. 

The study finds that family members were likely to share a loyalty to a particular brand, implying that both Mom and Dad have a preference for, say, Folgers Coffee. 

But when it came to price changes and promotional discounts, household members reacted differently, with some but not others heavily influenced by promos. The researchers argue that these results are consistent with the finding from University of Michigan’s Scott I. Rick, University of Pennsylvania’s Deborah A. Small, and Northwestern University’s Eli J. Finkel that suggests people in relationships don’t often share the same attitudes toward spending, and even have opposite tendencies at times. 

The researchers developed a model that describes the purchase choices of the households in their data. Their model sheds light on the concept of targeted marketing, prompting them to argue that analysts and marketers should consider not only household-level decisions but individual-level behavior. Because the model finds nuances associated with how family members respond to price drops and increases, Bruno, Cebollada, and Chintagunta advise marketers to understand the behavior of individuals in a household, saying the profits from targeted price promotions could increase by more than 50 percent as a result.