Customer Behavior Based Targeted Promotions as Predictors of Profitability in Repeat Online and In-Store Retail Transactions
Darla Moore School of Business at the University of South Carolina
Oracle Principal Investigator
This research collaboration will help Oracle Retail to leapfrog the competition in the fast growing customer pricing, revenue management and marketing analytics space. We expect this work to result in two or more publications and patents that will lead to Oracle Retail products which enable online and in-store retailers to predict and target the most effective promotions for their customers. Promotions are typically mailed/emailed/broadcast out to all customers indiscriminately in the hope of generating additional store/web traffic where a portion of these customers will buy a basket of goods, some of which are not discounted. One of the problems with this approach is that the promotion may be wasted on customers who would have bought the product/service anyway, resulting in a needless loss of margin. Another undesirable outcome is that the promotion induces a customer to only purchase the product/service that is discounted by the promotion. In the e-commerce world, this problem is even more complex given the assistance of web searching tools and on-line price checking and comparing capabilities. Meanwhile, research in the behavioral economics field has shown that customers are more sensitive to the market prices of certain items (e.g. gallon of milk) versus others (e.g. box of organic pasta), with the later presenting opportunities for retailers to achieve higher margins and thus, higher profits. What is needed, for both on-line and brick-and-mortar retailers, are promotions specifically targeted towards the customers who are induced to visit the retailer and/or its website (i.e. on-line store) in direct response to the promotion and who will also purchase a basket of goods containing items that are less market-price-sensitive. The methods summarized in this proposal represent cutting edge research areas intended to help both traditional brick-and-mortar and on-line retailers achieve this ideal in a computationally efficient manner. The EPI is actively involved in these areas and is one of the top five researchers in the world for the particular methods proposed here. The competitive landscape in this area includes firms such as SAS, PROS, DemandTec (acquired by IBM in 2011) and Predictix. Through his editorial roles for scientific journals and his consulting work with clients of these companies, the EPI is knowledgeable about many of the analytics solutions provided by these competitors and is proposing solutions that have the potential to distinguish Oracle Retail’s offerings as the most sophisticated and effective solutions within this space.