An Empirical Study of Software Sampling: Categorical Heterogeneity and Vendor Strategy

Abstract

Product sampling is a part of promotion tools to increase the sales of goods. Especially, the marginal sampling cost for consumer is close to zero in software market since consumers may download a freeware or trial version of the focal product easily through the internet. Unlike other consumable goods, sampling is an inexpensive strategy to software vendors. Hence, it is common practice in software markets. This research builds a multilevel model to illustrate categorical heterogeneity in software sampling and the effects of sampling strategy on sampling performance among software categories. Using the data collected from download.com, we find that: 1) sampling intensity is different among software categories; 2) trial version strategy outperforms freeware strategy in some categories, but on average its effect on sampling performance is lower than that of freeware strategy by 19%. These results serve to provide guidance for software vendors to choose the better sampling strategy.

Keywords: Software, Product Sampling, Freeware, Trialware, Bayesian Inference