The intensity of a crime prevention scheme can be measured in a variety of ways, including the number of measures implemented, the cost of the measures, the amount of staff time spent on project implementation, and levered-in-costs. The evaluation of the 21 burglary-reduction initiatives used a wide variety of datasets over the evaluation period (April 1999 through September 2001). These datasets included information on recorded crimes; social and demographic composition of the areas; intervention being implemented; the process by which implementation occurred; the costs of the project; activities in the area that might affect the crime rate; scheme inputs (levels of funding, staff time, purchase of locks, bolts, and alley-gates); and the outputs of the scheme (e.g., number of locks/gates fitted, number of crime-prevention packs distributed, etc.). Input intensity was calculated by dividing the total cost of the scheme by the number of households in the action area. Analyses that examined the utility of intensity measures in explaining changes in the burglary rate in scheme areas and the overall outcome of the scheme in terms of the number of burglaries prevented found that intensity was a key measure in explaining burglary reduction. There was a significant positive relationship between the overall scheme intensity and the number of burglaries prevented by the schemes. It was output intensity (number of measures actually installed) rather than input intensity (costs incurred by the project) that explained changes in the burglary rate in the scheme areas over time. As the intensity of the outputs increased, the burglary rate decreased, showing a direct relationship between the actual implementation of measures and reductions in the burglary rate. Further analysis also found that output intensity was a significant predictor of changes in the burglary rate even when other process issues that varied over time were considered. Taken together, the findings suggest that although the more successful schemes were the more intense ones, this did not mean they were the most cost-beneficial. This suggests there will be a trade-off when resourcing crime-prevention schemes. 6 figures, 3 tables, 14 references, and appended chart of standard outputs collected