This study aimed to investigate the effectiveness of metacognitive scaffolding strategy which is embedded in cooperative learning model in teaching writing to students at grade XI of SMAN 4 Kendari. In this study, quantitative data were used to examine the effectiveness of metacognitive scaffolding strategy on students’ writing, whereas qualitative data were used to describe the process involved in the teaching which applied the metacognitive scaffolding strategy. Three research questions were posed: (1) did the students taught under the metacognitive scaffolding strategy produce better writing than their counterparts who were not instructed in the same strategy?; (2) what were the writing scores gained by the students who learnt via the metacognitive scaffolding strategy?; (3) how was the metacognitive scaffolding strategy implemented in the process of teaching writing? The study was conducted at SMAN 4 Kendari to class XI students who enrolled in the 2015/2016 academic year. Total samples were 60 students which were split into an experimental class and control class. The samples were drawn using the convenient sampling technique. Instruments of the study included a writing test, interviews, and observation sheet. In the experimental class, a pre-test was administered to obtain data about students’ knowledge prior to treatment which applied the metacognitive scaffolding strategy in the teaching of writing, and a post-test was run to determine the students’ writing scores after receiving the treatment. In the control class, the teaching of writing applied a conventional method that was usually been used in the school. The students in this class also took a writing test, the results of which were used as a comparison to the experimental class. Additionally, some students and teachers were interviewed to obtain qualitative data and some observations were conducted during treatment in the experimental class. The results show that there was a significant difference between the pre-test scores and the post-test scores in the experimental class compared to the control class, with the students taught using the metacognitive scaffolding strategy gained higher scores than those who were not. Paired sample t-test was run to compare the pre- and post-test scores gained by both classes before and after treatments, whereas independent sample t-test was used to analyze the difference between gained scores recorded by the two classes. Furthermore, it was revealed that the process of teaching writing using the metacognitive scaffolding strategy has run well according to the lesson plan.