Product Bundling: Assortment Selection in Scale and Consumer Behavior
Abstract
Bundle promotions can take various forms in how retailers offer them to consumers (i.e., after a purchase incidence, as advertised deals, or personalized promotions) and how they are structured (e.g., buy one get one free, spend $100 get 25% off, or buy 3 get $20 off). They have become increasingly popular among brick-and-mortar and online retailers due to their potential for increasing revenue and profit. Designing an optimal set of bundle promotions in scale and understanding consumer response to these promotions are very relevant but challenging problems. In this talk, Ozge Sahin will share findings from two working papers on the topic.
In the first paper, the optimal design of bundle promotion assortments in scale is studied. The general bundle set selection problem is NP-hard. The researchers developed pseudo-polynomial and polynomial-time algorithms for special cases of the problem. They used insights from the special cases to develop a linear-time approximation with a performance guarantee under certain assumptions. The algorithms were tested in medium to large instances and showed close-to-optimal performance. Additionally, insights were presented on when bundle promotions are most useful to the firm by considering the interaction among products.
In the second part of the talk, empirical findings were shared on how bundle promotions affect consumer purchase and return behavior compared to simple markdowns. Using data from one of the largest apparel retailers in Turkey, the researchers found that bundle promotions not only increase the purchase incidence for certain categories but also decrease the return probability of products in the category, controlling for price, discount depth, and item characteristics. The returns of products offered with a bundle discount decreased on average by 10% to 24% for 20% to 40% bundle discounts compared to returns of the same products if purchased with a regular percentage discount. It was also shown that accounting for the impact of bundle discounts on consumer returns significantly changes the optimal promotion strategy for the firm and the products assigned to bundle promotions. This can translate to up to a 4% increase in the retailer's profits compared to a promotion strategy that does not account for returns.
Joint work with: Ali Fattahi, Sahar Hemmati, Wedad Elmaghraby
Bio
Ozge Sahin is a Professor of Operations Management and Business Analytics at the Johns Hopkins Carey Business School. She obtained her PhD and MS degrees in Operations Research from Columbia University. Her research interests include pricing, marketplace analytics with an emphasis on consumer behavior, and strategic capacity and supply chain management. Ozge has published papers in academic journals including Management Science, Operations Research, and Manufacturing and Service Operations Management, among others. She currently serves as the Department Editor of the Revenue Management and Pricing Area of Decision Sciences Journal and is an Associate Editor for Manufacturing and Service Operations Management and Operations Research. She is also an Amazon Scholar in the Pricing Research and Machine Learning group at Amazon.
