Nowadays the wireless sensor network (WSN) has been used for variety of applications and still lot of research in progress around the corner for the betterment of the wireless sensor network technology. In this paper, one such issues related to energy consumption in sensor node due to fixed sampling interval of sensing unit and its impact on redundant data is discussed with a possible solution. The association of sampling interval and its impact on energy dissipation in sensor node enforces the need for study on energy efficient adaptive sampling interval approach. The lack of serious consideration of outlier in sensor data degrades the performance of the existing adaptive sampling interval approach. The result of the proposed approach of in-network clustering algorithm shows the better efficiency towards detecting the outlier in real time. The results also showcase the better efficiency of proposed approach in terms of rapid optimization of sampling interval compared to simple variance based approach.