This research is a quasi-experimental with nonequivalent pretest-posttest control group design by using Concrete-Representational-Abstract (CRA) based on intuition approach and Conventional Learning (CL) conducted to investigate enhancement of students mathematical communication ability. Population of the research is seventh grade students in one of the Junior High School in Kampar district Riau Province, amounting to 175 students. The sampling research using purposive sampling techniques to choose two classes totaling 66 students who homogeneous characteristics. Before the experiment, the students are classified into high, medium, and low level of prior  mathematical ability. The instruments of research are daily test results and a mathematical communication ability test. The research found that the enhancement mathematical communication ability of students who received CRA based on intuition approach are better than those of students who recieved CL reviewed based on overall students, students of high level prior mathematical ability, students of medium level prior mathematical ability, and students of low level prior mathematical ability. Other findings, there are differences in enhancement mathematical communication ability of students between level of prior  mathematical ability (high, medium, and low) who received CRA based on intuition approach