Socio-Markal was a first attempt to take into account consumer's preferences through the modeling of a virtual technology directly in the energy model as a data set. This virtual technology would correspond to a marketing campaign for influencing people to consume less energy by changing their behavior or by choosing more efficient energy technologies. In this paper, we go a step further. We modify the structure of the energy model by including a Share of Choice model. The energy model which corresponds to a Mathematical Programming (MP) model is then transformed into a Mixed Integer Programming (MIP) model. The Share of Choice describes the consumers' preferences of a given energy service (e.g. lighting) through the optimization of “salient attributes” (i.e. elements of perceived value such as price, durability, ecological attributes). The data set or “part-worths” associated with the Share of Choice comes from a psychological statistical technique called “conjoint analysis”. We propose a simple case study about lighting bulbs just for concept proven purposes. We are currently conducting a survey in Romania related to household appliances. This contribution should help policy makers to better target awareness campaigns when demand-side consumption patterns and technology purchases are not grounded on energy and environmental efficiency criteria.