We present a novel approach to identify individuals’ preferences in the context of sustainable energy consumption by letting them play a public good game (PGG). The study will be conducted using an energy-efficiency website developed by us and operated by an Austrian utility company that currently facilitates 9,929 users out of which 1,000 will be randomly selected as participants. Laboratory experiments on PGGs identified two types of cooperative behavior: free riders and conditional cooperators. While free riders tend to act unaffected by the behavior of others, conditional cooperators are stronger influenced by the perception of their peers. In our study, participants receive normative feedback on their consumption, and we investigate the moderating effects of the PGG’s results on changes in energy consumption. The findings will contribute to the design of information systems to promote environmental sustainability because their effectiveness is increased if the consumption feedback provided reflects the recipient’s preference.