A randomized pricing decision support system in electronic commerce

The Internet has provided great convenience for online shoppers and has presented unprecedented opportunities for online retailers to understand their customers. Getting the pricing right has emerged as one of the ultimate keys to the success of electronic commerce. Although some online retailers have tried some personalized pricing strategies for perishable capacity or inventory in some industries, consumers' resistance to price discrimination is still a great concern. Can we develop other price discrimination strategies for online sellers to sell standard durable products without giving the impression that they are treating their customers unfairly? Randomized pricing, which is proposed in this paper, belongs to this kind of strategy. In this paper, we present a framework that can be used to study the randomized pricing strategy by incorporating some new features into electronic commerce. For example, information asymmetry about the prices of products does not exist across internet users because of easy access to price information and very low searching cost. Consumers' reneging behavior is also considered. Online consumers usually wait up to a certain period of time for deals. Specifically, we model online retailers' price variation as a Markov process in which the price randomly switches between high level and low level. Strategic consumers make a tradeoff between buying immediately at a high price with instant utility or buying later at a low price with a probability and discounted utility. We show in this paper that randomized pricing strategy can always generate more profit than flat pricing strategy. The effects of consumers' patience and discount factor on optimal prices and promotion probability are studied. Finally, we show that the optimal benefit that the retailer can obtain from hiding promotion probability depends on the value of the discount factor.