The fact that occupant behaviour has a large effect on buildings' energy consumption is well accepted. Occupants affect energy consumption by their behaviours: using lighting, appliances, thermostats etc. and interacting with envelope components such as windows and blinds. During last decades, specific ICT solutions have been developed for addressing behavioural changes toward energy saving and consequently leading to energy conscious occupant behaviours. This paper presents a mean of interaction between sensors and users for tackling behavioural changes toward energy saving in homes. This is based on premade rules and instructions (referred to as recipes), meant for tackling energy hungry everyday life actions. Moreover, it investigates the use of nonintrusive appliance load monitoring (NIALM) system. Among all the available off the shelfs sensors, NIALM seems to be very promising, especially if combined with self-learning algorithms, for detecting energy consumption of electrical appliances. This combination can potentially solve existing constrains: detecting and identifying appliances with very small or continuous electricity consumption or that turn on or turn off slowly. Literature information have been cross checked with home energy consumption monitored data in order to create pre-formulated energy recipes for empowering occupants addressing specific energy hungry behaviours. Besides establishing virtuous occupant behaviours, devoted to a rational and judicious use of energy, the adoption of energy recipes aims to generate critical knowledge about the intrinsic meaning of technical variables.