Problem-solving and mathematical communication are essential skills needed by students in learning mathematics. However, empirical evidence reports that students' skills are less satisfying. Thus, this study aims to improve students' problem-solving and mathematical communication skills using a Metacognitive-Based Contextual Learning (MBCL) model. A quasi-experimental non-equivalent control group design was used in this study. The participants were 204 fifth-grade students; consisting of experimental (n = 102) and control (n = 102) groups selected using convenience sampling. This study was conducted in four Indonesian elementary schools in the first semester of the academic year 2019/2020. The Problem-Solving Skills Test (PSST) and Mathematical Communication Skills Test (MCST) were used as pre- and post-tests. In order to analyze the data, one-way ANOVA was used at the 0.05 significance level. The results showed that students in the experimental group had higher post-test scores than the control group in terms of problem-solving and mathematical communication skills. It can be concluded that the MBCL effectively promotes fifth-grade students' problem-solving and mathematical communication skills. Therefore, it is suggested that MBCL should be used more frequently in primary school mathematics to further improve students' problem-solving and mathematical communication skills.