This study aims to examine the effect of implementing differentiated instruction within the Discovery Learning model on students’ communication skills and learning outcomes. This research employed a quasi-experimental design with a posttest-only control group, involving 55 eleventh-grade students at SMAN 2 Bantaeng in the 2025/2026 academic year. The experimental class received differentiated Discovery Learning, while the control class was taught using conventional Discovery Learning. Data were collected through observation (communication skills) and essay tests (learning outcomes), and analyzed using descriptive and inferential statistics, including MANOVA. The results showed that students in the experimental class achieved higher communication skills than those in the control class. The mean score of oral communication in the experimental class (89.87) was higher than in the control class (77.08), while written communication scores were 87.99 and 75.38, respectively. In addition, 51.72% of students in the experimental class were categorized as very skilled, whereas no students in the control class reached this category. Learning outcomes also showed significant differences, with the experimental class achieving a higher average score (85.14) compared to the control class (75.85). Most students in the experimental class were in the high (75.86%) and very high (13.79%) categories, while the control class was dominated by the moderate category (84.62%). These findings indicate that the integration of differentiated instruction within the Discovery Learning model is effective in improving students’ communication skills and learning outcomes. This approach supports more equitable learning, accommodates diverse student needs, and promotes the development of essential 21st-century skills such as critical thinking, collaboration, and communication.