The purpose of this study was to obtain accurate information about one of the efforts to improve students' understanding of concepts in Indonesian lessons, especially describing objects. This action research process is planned for three cycles, where each cycle will be carried out in accordance with the changes to be achieved such as what has been designed in the factors according to the procedure: planning (planning), action (acting), observation (observing), and reflection (reflecting). ) The researcher is a Class V teacher at MIN 11 Aceh Utara collaborating with colleagues who are considered capable and competent in the context of this research. The subjects of this study were students of Class V MIN 11 North Aceh in the 2017/2018 academic year, involving all students of Class V-A. Data was collected using instruments in the form of questionnaires, tests, question sheets, and observation sheets. Questionnaires and tests were conducted on students to see the ability to understand concepts in Indonesian language lessons about describing objects before and after being remedial by peer tutors. Student question sheets and observation sheets were used to see the implementation. The data that has been collected in the form of quantitative data, is processed and analyzed through the stages of data reduction, exposure, and inference. The results of this study stated that understanding the Indonesian language concept of describing objects through remedial activities by peer factors in small groups could increase students' test scores. This can be seen from the behavioral observation table during remedial activities by peer tutors as well as the results of the acquisition of grades in each cycle. Marked on the indicator of understanding Indonesian concepts about describing objects in the first cycle, students only actively discussed with peer tutors, only ± 3 people from each group or around 37.50%, but in the second cycle there was an increase so that on average each group actively discussed There are 6 people in each group. Meanwhile, a significant increase occurred in student activity in answering questions given by peer tutors from 3 people or 37.50% in the first cycle to 56.67%.