Industries are increasingly demanded to became more green and, at the same time, the process of becoming more connected and automated enabled the measurement of actual energy flows. Indeed, the use of power meters is increased and energy related KPIs are becoming more relevant. High industrial interest is pointed to monitor the energy consumption, to automatically assess energy efficiency, and to identify potentialities for improvement. Nevertheless, monitoring systems are often limited to a certain area or resource and the plant-level vision is not common. Also, the attention is often focused on a particular process, usually energy-intense. The work presents the application of an on-line energy monitoring system to the analysis of the overall plant consumption of a real factory in Italy. The factory produces a full range of valves and actuators for the most critical applications of Oil & Gas industry, and faces with high energy consumption due to the nature of the involved processes (i.e., welding, machining, painting, and high-pressure testing). Monitored data are used to understand plant hot-spots in terms of energy consumption and to create a map of the most (or least) efficient areas of the plant so that specific measures can be implemented to reduce carbon footprint. Indications for a data-driven monitoring system to enable the identification of unexpected and unwanted behaviors are defined.