PurposeThis research aimed to examine the impact of input, process, output, productivity and outcome variables on university research performance and the indicators that represent them in order to improve academic quality and contribute to government policy.Design/methodology/approachThe quantitative approach was used through a survey method that obtained samples using questionnaires from 150 leaders of research institutions and continued analysis using the structural equation modeling-partial least square (SEM-PLS) to test the developed model.FindingsExcept for the relationship between process and productivity variables, all variable relationships had a positive and significant effect. Furthermore, the input, process, output, productivity and outcome variables each include seven, twelve, four and ten indicators.Research limitations/implicationsThis study has several ramifications because it provides a clear policy input and advances science. As a prelude to developing research performance assessment tools that take into account variances in a tertiary institution, this research aids in the implementation of national policies for assessing research performance in postsecondary institutions.Originality/valueTo improve the accuracy of the information acquired, we conducted a survey among the heads of research units at various higher-ranking Indonesian universities, taking into consideration their skill and experience in leading research organizations and conducting research. Other than that, our belief in the originality of our manuscript is strengthened by the way we applied systems theory to construct a performance evaluation model that examines each contribution made by each system aspect.