The main problem with previous evaluation studies on situational crime prevention programs is that very few of them have been carried out under controlled, experimental conditions. Prior evaluations have produced contradictory results, been subject to a lack of subject compliance, have been contaminated by the use of multi-tactic interventions and publicity campaigns, and have either lacked control groups or equivalent comparison groups. The author argues that a place-based randomized experiment in burglary reduction is well suited to Denmark and should be conducted to contribute to the dearth of knowledge in this area. The author outlines an experimental design in which the unit of analysis is the individual household dwelling rather than the neighborhoods, police beats, or cities that have been the unit of analysis in previous research. By making household dwellings the unit of analysis, dwellings can be randomly assigned to treatment and control groups and post-treatment differences in burglary levels can be observed. Denmark proves a particularly fertile ground for such experimentation because of the quality of the Dutch registries on dwellings and the fact that residential burglaries are relatively well reported to police in Denmark. Tables, footnotes, references