This study aims to describe the needs of Android-based learning media on the digestive system material, design Android-based learning media for fifth grade students, analyze the effectiveness of Android-based learning media to improve learning outcomes of fifth grade students' digestive system materials. This research is a research development or Research and Development (R&D) using the Borg and Gall model development method which aims to develop Android-Based Learning media. Qualitative data collection techniques through documentation, interviews and observations, while quantitative data collection techniques through tests and questionnaires. The resulting product is in the form of Android-Based Learning. Analysis of Android-Based Learning needs by teachers and students, validation is carried out by material experts, media experts and practitioners, effectiveness test by analysis of learning outcomes. The subject of the limited product trial in this development research was class V SD Negeri 1 Nambuhan, namely 17 students. The test subjects were 54 students as the experimental class and 47 students as the control class. Data collection uses an Android-Based Learning needs questionnaire, Android-Based Learning assessment, and student learning outcomes tests. The data analysis technique uses Android-Based Learning needs analysis, validation sheets, and t-test acquisition analysis to determine the effectiveness of the Android-Based Learning developed. The result of the research is the development of Android-Based Learning based on the needs of teachers and students and has received validation from experts. Product trials are carried out on a limited scale and on a wide scale. The results of the limited trial showed an increase in learning outcomes with an n-gain value of 0.696 in the medium category. The results of the large-scale trial on the pretest with a significance value of 0.559 which means there is no difference in the average value in the experimental class and the control class. The posttest value with a significance value of 0.000< 0,05 means that there is a difference between the average posttest value of the control class and the experimental class. From the statistical test, it can be concluded that the development of Android-Based Learning media can improve learning outcomes in the digestive system material for fifth grade elementary school students in Purwodadi District.