The purpose of this study was to; analyze the differences between Gusjigang learning approach and conventional learning on student learning outcomes. The type of research used was a quantitative research method with a nonequivalent control group design. The population consisted of all fourth-grade students of Madrasah Ibtidaiyah TBS Kudus Regency. The sample was selected through a purposive sampling technique, so that class IV B was selected with 23 students as the experimental group and class IV C with 23 students as the control group. The data collection technique uses test techniques, namely the pre-test and post-test questions in the form of multiple choices and brief descriptions. The results obtained in the experimental class, namely the class that in the experimental class that experienced learning with the local wisdom model, Gusjigang gained an increase in learning outcomes by 83 in the high category. The N-Gain results in the control class are students who are taught with conventional learning, students who get low n-gain results are 56%, students who get moderate N-gain results are 44%, and none of the students achieves n-gain results high. Whereas in the experimental class, the students who were taught using Gusjigang's local wisdom-based learning, students who got a low n-gain result of 9%, students who got a moderate n-gain of 56%, and students who achieved a high n-gain of 35 %. This shows that the acquisition of student learning outcomes in the experimental class is better than that of the control class.