In recent years, there has been a growing interest in assessing the level of performance and eciency of the foodservice industry. As a result, an increasing number of studies applied eciency frontiers methods to quantify rm (in)eciency. Starting from the benchmarking studies on restaurant eciency, this paper aims to develop a meta-analysis based on 77 observations from 25 studies published in scientic journals from 1998 to 2020. The estimated eect size in our meta-analysis is equal to 0.842 and it is statistically dierent from zero, while the Cochran's Q test for the heterogeneity in our sample hints at the absence of heterogeneity in the previous studies on restaurant eciency. A meta-regression analysis partially supports this result but also highlights the importance of assuming appropriate return to scale, given the peculiarity of the sector.