In this paper, we propose a new system aiming to support government agency decision makers to design new services tailored to citizen profiles in a complex and distributed e-government scenario. Specifically, our system assists government agency managers, who plan to activate new services, to identify those citizens who could gain the highest benefit from each of them. Managers can, then, exploit this information to decide what services should be activated and how they can be tailored to citizen needs and desires. Our system can handle more government agencies and a great number of citizens simultaneously; as a consequence, it is well suited for a complex e-government scenario. This paper first illustrates the proposed system; after this, it reports various experimental results; finally, it presents a comparison between our system and other related ones already presented in the literature.