Certain sensor applications can tolerate sensing data being delivered within certain delay. In such a case, the mobility of sink node can be exploited in conjunction with sensing data buffering to greatly enhance the energy efficiency of data gathering in wireless sensor networks. That is, each sensor node sends its data to a location (called 'sink site') over the migration path of the mobile sink and the sensor node at the location stores the data until the mobile sink node arrives. In this paper, we propose a heuristic algorithm that determines the most effective sink site from the energy efficiency view point for each sensor node and determines the route between the sensor node and the corresponding sink site. Our heuristic is designed on the basis of the insight obtained from the analysis of the optimal solution. The proposed scheme requires only local knowledge, which ensures scalability. The simulation results show that the proposed heuristic algorithm results near-optimal performance under various scenarios.