The implementation of group investigation by paying attention to the students' input quality is believed to be able to create a fun learning process and be able to reach competency units under performance criteria. The study aims to explain the students' input quality and to explain the role of group investigation in accommodating the input quality to students' competency improvement. This study uses two-factor measurement with a factorial version of the non-equivalent pretest-posttest control group design. The variables in this study are competency, group investigation, and students' input quality. The total samples are 116 students and each treatment decided 25 subjects as an analysis unit. The results of the study are the students' input quality varies in the accounting cycle, especially to process some information related to the closing process and reversing process, and the group investigation model accommodates input quality in improving students' competency.