When does it pay to invest in pricing algorithms?

Abstract

Nowadays, firms frequently use big data and pricing algorithms to offer consumers personalized prices according to their willingness to pay. Advances in information technologies have further facilitated the use of customized pricing, which we consider an important facet of transformative marketing. At the outset, personalized pricing may appear to reduce the asymmetry of information between the firm and consumers, and benefits the firm but hurts consumers. To investigate this view, we consider a novel setting in which consumers must incur search costs to make an informed purchase from the firm. Contrary to conventional wisdom, we find that personalized pricing can sometimes make both the firm and consumers better off, thus leading to a win-win situation. We also show that an imperfect pricing algorithm can outperform a perfect one, thereby explaining why certain retailers like Amazon are adopting imperfect pricing algorithms. On the one hand, a moderately reliable pricing algorithm gives high-preference consumers a chance to be misclassified as low-preference consumers and obtain a low price, thereby encouraging consumer search. On the other hand, a highly reliable pricing algorithm significantly reduces consumers' surplus, which stifles consumer search. As a result, both firm profit and consumer surplus can be non-monotone in the reliability of the algorithm. To the best of our knowledge, this is the first paper that documents the effect of personalized pricing under consumer search.