Machine‐type communications (MTC) are a rapidly growing technology, which is expected to generate significant revenues to mobile network operators. In particular, smart grid is predicted to become one of the key MTC use cases that involves unattended meters autonomously reporting information to a grid infrastructure. With this research, we consider a typical smart metering MTC application scenario in the context of 3GPP LTE‐advanced wireless cellular system featuring a large number of devices connecting to the network near‐simultaneously. The resulting overload of the random access channel requires a novel evaluation methodology based on comprehensive analysis and simulations. In this paper, we target to complement a validated evaluation framework fully compatible with the 3GPP test cases with a thorough analysis of random access channel performance in overloaded MTC scenarios. We also look at the regular MTC operation, when the devices are sending their data after initial network entry has been performed. By including energy consumption into our methodology together with the conventional performance metrics, we aim at providing a complete and unified insight into MTC device operation, including its energy efficiency. Copyright © 2013 John Wiley & Sons, Ltd.