Summary This article investigates the event‐triggered Kalman consensus filtering (ET‐KCF) problem for distributed sensor networks with intermittent observations. First, a novel ET consensus filtering structure is designed for sensor networks with intermittent observations. With the proposed consensus filtering structure, a new ET mechanism that is more efficient than the existed ones is designed to schedule transmissions of local estimates. Then, an optimal ET‐KCF in the sense of minimum mean‐square error is developed. For reducing the computational complexity of filtering algorithm, a suboptimal ET‐KCF is further proposed. Moreover, the stability of the suboptimal ET‐KCF is analyzed. Simulation results verify the validity and superiority of the proposed method.