This paper analyzes music charts of an online music distributor. In music charts, the digital music provider displays a daily ranking of 1st ~ 100th and a weekly ranking of 1st ~ 1,000th songs on its website. And the ranking of each song is assigned based on streaming volumes and download volumes. This paper studies how the online music distributor should set its ranking policy to maximize the value of online music ranking service. Compared to the current ranking mechanism which is being used by music sites and only considers streaming and download volumes, a new ranking mechanism is proposed in this paper. A key improvement of the new ranking mechanism is to reflect a more accurate preference pertinent to popularity, pricing policy and slot effect based on exponential decay model for online users. A ranking model is built to verify correlations between two service volumes and popularity, pricing policy, and slot effect. An empirical analysis is followed to illustrate some of the general features of online music charts and to validate the assumptions used in the new ranking model. The results from the empirical work show that the new ranking mechanism proposed will be more effective than the former one in several aspects.