A first component of a process control loop (e.g., a controller or I/O gateway) monitors for and detects performance degradation of a second component of the loop by sending heartbeat messages to the second component via a diagnostic channel different from a control communications channel via which the first and second components communicate control messages for controlling an industrial process. The second component utilizes its control message interpreter to return received heartbeat messages to the first component via the diagnostic channel. The first component detects degradation of the second component when the round trip time (RTT) of a heartbeat message falls outside of an acceptable range of RTTs for the second component, and may suggest or automatically initiate mitigating actions. The first component may determine the average RTT or expected response time of the second component and acceptable range of variations based on a sample number of measured RTTs.