Praveen Kopalle is professor of marketing and chair of the marketing area at the Tuck School of Business at Dartmouth College. His research interests include pricing and promotions, new products and innovation, customer expectations and e-commerce.
Kopalle has won many international awards. Most recently, he and his and co-authors P.K. Kannan, Lin Bao Boldt and Neeraj Arora won the 2014 Davidson Award for the best paper published in The Journal of Retailing in 2012. The paper, "The Impact of Household Level Heterogeneity in Reference Price Effects on Optimal Retailer Pricing Policies," received the most votes from a 60 member award committee. Shankar Ganesan, a professor at the Mendoza College of Business at Notre Dame andEditor-in-Chief of the Journal of Retailing, said the paper was selected for its "originality, technical competence and a strong contribution to the theory and practice of retailing."
I had the opportunity to interview Kopalle:
Congratulations on the Davidson Award. Please tell me about the findings in your study.
Thank you. I was really thrilled to receive this award because it was unexpected it, and it is one of my favorite papers.
My team and I used retailer data collected from loyalty card purchases to understand consumer reactions to price changes and fluctuations between in-store prices and the reference prices they carry in their heads. Leveraging the consumer behavior information we collected, overall category profitability can increase by as much as 76 percent.
What is "household-level heterogeneity"?
"Household-level heterogeneity" is a marketing research term which simply means that different households behave differently. One behavior that retailers have long desired to better understand is how individual households adjust their buying when store prices go up or down. For most of retail history, stores had to assume all their customers reacted similarly to price changes. But with the advent of loyalty programs, stores began collecting valuable data that could connect their customers’ shopping patterns to changes in prices.
Analyzed correctly, this data shows consumer price sensitivity – the point at which a change in price causes a consumer to alter purchasing behavior – and how that sensitivity impacts consumer "reference price" or the preconceived notion of the price of a certain item. These reference prices change over time and are a function of past prices.
What are reference prices?
Reference prices are anchors or guides that households compare with the observed purchase price of a product. If the observed price is greater than the reference price, it is perceived as a "loss." If it is smaller than the reference price, it is perceived as a "gain."
These perceptions of loss and gains can have significant impact on the purchase probabilities of the brands by households. The references prices can vary significantly across households and determine pricing policies of retailers.
As an example, how do soda sales fare when prices change?
Say people are used to buying Coca Cola (Coke) at $2, i.e., their reference price is $2. However, it turns out that more people are sensitive to a $0.25 price decrease around this reference price relative to a $0.25 price increase. Because of such "gain-seeking" consumers, reducing the price of Coke to $1.75 during a week would increase demand for Coke by 15 percent; this incremental bump could be due to consumer stockpiling as well as shifting from other lower priced brands that they would have otherwise bought.
The following week, when Coke increases its price to $2.25, the decrease in demand is only 10 percent since these consumers are less sensitive to a price increase around the reference price. This means that when Coke increased its price, it didn't lose that many of these customers. The bottom line is that retailers can now figure out, in each category, the proportion of its customer base that is more (or less) sensitive to price changes around the reference price and then determine the sweet spot for the prices of its products in those categories.
Do generic brands steal market share when their prices are discounted?
What we find in this regard is quite interesting. That is, there is an asymmetric effect: When the national cola brands run a promotion they steal more share from the generic brands relative to the amount of market share that the generics can garner from the national brands when the generics go on sale. Essentially, the "clout" of national brands on the generics appears to be stronger than their "vulnerability" to the generic brands.
How should retailers plan their pricing strategies for periods longer than two weeks?
The beauty of our modeling framework is that it works for any time horizon that a retailer chooses. The number of periods (or weeks) are determined by the retailer’s calendar and it typically turns out to be thirteen weeks. In our algorithm, we can set whatever planning horizon that a retailer would prefer.
Once the number of periods is set by the retailer, we can then optimize the prices of the various brands in a category during those periods. Another nice feature of our model is that as we move through time, we obtain more sales data, and this would help us refine our consumer price and reference price sensitivities as well as update our optimized prices in the future periods.
Lisa Chau is the founder of Alpha Vert, a private consultancy focused on social media and cross–platform marketing. Previously, she spent five years working for her alma mater Dartmouth College, as assistant director of alumni affairs and assistant director of PR for the Tuck School of Business. She has also taught at MIT, and guest lectured MBA and undergraduate courses in e-business strategy at Baruch College and The New School.