Price and revenue bounds for bundles of information goods

Abstract

In this paper, we investigate the behavior of the expected revenue function generated from selling bundles with arbitrarily many components. A motivating example of such bundles includes the production and delivery of digital content, where variable costs are generally negligible. Specifically, we derive generic lower and upper bounds for the expected revenue function even when accounting for arbitrary, potentially complex, dependence structures among the bundle components. The expected revenue bounds in turn provide upper and lower bounds regarding the optimal pure bundle price. Our results reconcile the extant bundling literature involving expected revenue bounds, while sharpening some of these results even when relaxing the traditional assumption of independence among the valuations for the bundle components. We show how these bounds can be further tightened when the seller has additional information about the dependence relationship. Since these results effectively reduce the search space for the optimal bundle price, the pricing bounds provided by our framework have important managerial implications.