Dynamic Pricing and Customer Relationships: Learning by Selling

Knowing each customer’s preferences enhances the ability to design and market the right offer in terms of content and price. The more a retailer interacts with its customers, the more it learns as it observes revealed preferences. Lower prices typically lead to more interactions and more learning but lower short-term profits. This paper brings the concept of learning functions to this dynamic pricing context to analyze how firms can quantify the trade-off between the short-term profits and learning so as to maximize customer lifetime value. The pricing problem is formulated as a dynamic program and our main finding is that the shape of the learning curve and the firm’s position in the learning curve significantly affect the optimal pricing policy. Our analysis show that simple heuristics such as a twoprice policy can significantly improve profits when compared with the predominant myopic pricing paradigm.