Abstrak
This study investigates the effectiveness of using DeepL as a learning media to improve students’ descriptive writing ability in an Indonesian secondary school context. A quantitative approach with a quasi-experimental design was employed, involving an experimental group (17 students) and a control group (14 students). Both groups were administered a pre-test and a post-test to measure writing performance. The experimental group received descriptive writing instruction integrated with DeepL as a guided learning tool, while the control group was taught using conventional instruction. Data were analyzed using descriptive statistics and an independent samples t-test. The results revealed a statistically significant difference between the post-test scores of the experimental and control groups (p = 0.028 0.05). Although the magnitude of the effect was relatively small, the findings suggest that DeepL, when used pedagogically under teacher guidance, can function as a digital scaffolding tool to support vocabulary use, grammatical accuracy, and text organization in EFL writing classrooms.