Industry uses 'benchmarking' as a method of organisational comparison. Performance levels, benchmarks, for an industry group or successful firm are identified and these are used to identify best practice. At a simplistic level the comparison can be in terms of outputs. Clearly a more comprehensive approach requires significant evaluation of the production procedures which means that the focus is on the process by which inputs are converted to outputs. Despite the many problems associated with mapping industry-based jargon and procedures on to education, there is a trend, promoted by the federal government, to consider benchmarking in schools in Australia. Despite the potential for abuse of such there does seem some value in a considered benchmarking exercise, i.e., having general data on how large numbers of students perform in certain tasks, the types of strategies they employ, the types of errors they make and, where possible, linking such findings to a theoretical position concerning cognitive development. This paper reflects on the issue of benchmarking, and in addition, provides information on a large benchmarking exercise, undertaken with support from the NSW Board of Studies, on Year 12 student performance in mathematics as indicated by an analysis of students' examination scripts.