A key factor to energy-efficiency of heating in buildings is the behavior of households, in particular how they ventilate rooms. Energy demand can be reduced by behavioral change; devices can support this by giving feedback to consumers on their behavior. One such feedback device, called the ‘CO2 meter’, shows indoor air-quality in the colors of a traffic light to motivate so called ‘shock ventilation’, which is energy-efficient ventilation behavior. The following effects of the ‘CO2 meter’ are analyzed: (1) the effect of the device on ventilation behavior within households, (2) the diffusion of ‘CO2 meter’ to other households, and (3) the diffusion of changed behavior to households that do not adopt a ‘CO2 meter’. An agent-based model of these processes for the city of Bottrop (Germany) was developed using a variety of data sources. The model shows that the ‘CO2 meter’ would increase adoption of energy-efficient ventilation by c. 12% and reduce heating demand by c. 1% within 15 years. Technology diffusion was found to explain at least c. 54% of the estimated energy savings; behavior diffusion explains up to 46%. These findings indicate that the ‘CO2 meter’ is an interesting low-cost solution to increase the energy-efficiency in residential heating.