To increase the share of intermittent renewable energy in our production mix, occupants of buildings can be called upon to lower, anticipate or postpone their consumption according to the network balance. This article presents a small-scale field experiment aimed at introducing demand response in the workplace. We test the impact of load-shedding signals assorted with incentives on the energy consumption of workers in the tertiary sector. Two incentive schemes are tested, namely, an honorary contest and a monetary tournament. The results show a reduction in workers power demand during the load-shedding periods when the incentives are based on the honorary contest. In contrast, the monetary tournament where workers can win money according to their behavior seems to have no impact. The results also suggest that few workers can be responsible for a large part of energy consumption while the building is partially automatically controlled.