In order to optimize energy consumption in smart grid, demand side management has gained a lot of attention recently. While existing research works attempt to optimize energy consumption either from the view point of the power company or that of users, we investigate whether it is possible to consider both parties' interests at the same time. In this paper, we propose a novel energy price model, which is a function of the total energy consumption in the considered system. In addition, a new objective function, to optimize the difference between the value and cost of energy, is proposed. The power company sends the energy price parameter and the latest consumption summary vector information to the users sequentially. Upon receiving these information, a user can optimize his own schedule and report it to the company. The company then updates its energy price parameter before communicating with the next customers. A two-step centralized game is proposed that models this interaction between the power company and its consumers. The game aims at reducing the system peak-to-average power ratio by simultaneously optimizing users' energy schedules and lowering the overall energy consumption in the system. Through simulation results the performance of the proposed game-based demand side management technique is evaluated.