The approach of studying a form of practical fire safety measure by considering the population of buildings to which it is applied has led to some interesting quantitative results. These include the distribution of the population of detection equipment and some indication of its possible effectiveness. On the basis of the model presented here the current population of detectors should be present at the start of about 4800 fire incidents, or 15% of the total, in non-domestic buildings in the UK. This then leads to some indication of the performance requirements both in terms of reliability and acceptable false alarm rate. Statistical studies of this nature ought therefore to be useful to designers and legislators when considering new products and standards. Further work should be done to improve the accuracy of some of the statistics used as inputs.