This paper is related to the phenomenon puzzling unduly low take-up rate for social assistance benefits in France. In order to investigate this phenomenon, we conduct an experimental evaluation in the form of a randomized control trial involving the influences of informational availability and complexity. We examine the impact of a change in the information set which is sent to these households just after they claim the benefit, seeking a behavioral response to this particular ‘nudge’. Our findings suggest that a costless action on the part of program administrators is able to substantially increase take-up for certain types of beneficiaries. In order to be effective, these actions should target households according to their individual attributes.