The background of this research is the condition in the field that there is still a lack of digital teaching materials for Arabic language learning. So far, teachers only use textbooks that are still monotonous in their implementation, not dynamic enough to see the development of technology which is currently very rapid. The objectives in this study are (1) to determine the development of digital-based Arabic teaching materials with the Contextual Teaching and Learning approach, (2) to determine the effectiveness of using digital-based Arabic teaching materials with the Contextual Teaching and Learning approach. This research method is mixed method with Research and Development (R&D) design using Borg and Gall model. The results of this study showed a difference, the average pretest score was 38.9 and the average posttest score was 86.4. From the results of the pretest and posttest, the results of the normality test were pretest 0.088>0.05 and posttest 0.151>0.05, so the pretest and posttest data were normally distributed. Then the results of the homogeneity test show the sig value> 0.05, namely 0.571> 0.05, so the data variance is homogeneous data. From the paired t test it is known that the sig value. (2-tailed) < than 0.05, namely 0.000 < 0.05, so there is a real difference and influence between learning outcomes on pretest and posttest data. N-Gain Score test results are >78.6108 then the product development of digital-based Arabic language materials with contextual teaching and learning approach gets a very effective category.