Does popularity decide rankings or do rankings decide popularity? An investigation of ranking mechanism design

摘要

We analyze the music charts of an online digital music distributor that displays real time and weekly rankings on its website, and study how ranking policy should be set to maximize the value of its online music ranking service. The existing mechanism considers only streaming and download volumes, while the new ranking mechanism reflects more accurate preferences for popularity, pricing policy, and the slot effect based on the exponential decay of attention. The new ranking model is designed to verify correlations with two kinds of service volumes for popularity, pricing policy, and the slot effect. Slot mechanism design is analyzed in an heuristic way. Our analysis shows that music content sellers maximize benefits by assigning their own music items to the highest-ranking slot, which provides visibility. Also sellers can strategically design the slot size to influence the popularity of music items. Music content buyers gain indirect benefits by getting segmented ranking slots and reducing search costs. Empirical analysis illustrates the features of the online music industry and validates hypotheses constructed around the new ranking model. The results show that the new ranking mechanism is more effective.