\ Research on digital media in English literacy has largely emphasized technological effectiveness while overlooking the role of learners’ sociocultural context, particularly in non-urban settings. This study develops and evaluates a contextual Digital Comic-Based Approach (DCBA) designed for junior high school students in a rural coastal area of Southeast Sulawesi. Using a mixed-methods explanatory sequential design, the study involved 13 eighth-grade students and combined pre- and post-tests with semi-structured interviews. Quantitative analysis indicated an increase in students’ mean literacy scores from 33.08 to 41.54, with a mean difference of 8.46 points. A paired-samples t-test showed that the improvement approached statistical significance, t(12) = –2.13, p = .055, with a moderate effect size (d = 0.59). Qualitative findings revealed that students relied primarily on visual elements such as images, layout, and character expressions as central semiotic resources for constructing meaning, enabling them to compensate for limited linguistic proficiency. Engagement emerged from the coordinated orchestration of visual and textual modes grounded in students’ coastal lived experiences, which reduced cognitive barriers and supported vocabulary and pronunciation development through contextualized input. These findings suggest that the pedagogical value of digital comics in under-resourced EFL contexts lies less in immediate statistical gains and more in their capacity to mediate meaning-making through contextual multimodal design. The study contributes to multimodal literacy research by demonstrating how socially situated visual orchestration can expand textual accessibility for beginner learners and offers a context-responsive model for digital material development in rural EFL education.