An ongoing debate in the literature on efficiency of higher education institutions concerns the indicator for research output in the empirical analysis. While several studies chose to use the number of publications, others rely on the overall amount of research grants. The present study investigates whether both lead to similar or different assessments of universities. Besides the amount of research grants and the absolute number of publications, the number of publications belonging to the 10% and 1% most frequently cited papers in the corresponding subject category and publication year are evaluated. We show that there is a high correlation of efficiency values between the estimations using these indicators; however, the concordance is partly lower. The results do not only provide a helpful guideline for researchers, but are also valuable for policy makers deciding which incentives to create through funding.