Purpose: The purpose of this paper is to measure and compare the performance of Portuguese secondary schools. Design/methodology/approach: Some data on the schools were collected via an electronic questionnaire sent to 103 secondary schools of the centre region of Portugal; other needed data were available through the Centre Regional Education Authority. Of the 33 schools that participated in the study, there was complete data for 29. A non-parametric technique, data envelopment analysis (DEA), was used to assess the sample of schools and the results obtained were compared with some preliminary results of a national evaluation programme of schools. Findings: From the findings, the tentative conclusion is that most schools are following national education policy priorities relating to decreasing dropout rates and increasing completion rates. School performance does not seem to relate to geographic location, size of the school, typology of the school or rotation of its executive committee. The paper concludes by discussing the need for metric benchmarking exercises of the type proposed, to inform schools, evaluators and policy decision makers. Finally, the complementarities between metric and practice benchmarking exercises are argued for. Research limitations/implications: The paper reports on an exploratory study, and thus it contains limitations which need to be overcome by further work. The sample used is small and self-selected; the DEA models used are quite simple; also, it would be desirable to compare the results of the DEA, with parametric techniques that have been used to evaluate school performance. Originality/value: The paper makes a contribution to the understanding of schools' performance management, in order to inform schools, evaluators and policy decision makers. (Contains 1 figure and 11 tables.)