The objectives of this study are to find out whether or not there is a significant improvement before and after the treatment on the eleventh grade students’ reading comprehension scores of SMA Karya Ibu Palembang by using Feature Walks strategy and to find out whether or not there is a significant difference on the eleventh grade students’ reading comprehension scores of SMA Karya Ibu Palembang between the students who are taught by using Feature Walks strategy and those who are not. In this study, the researcher used Quasi Experimental Design using pretest-posttest nonequivalent groups design. There were 60 students taken as sample. Each class consisted of 30 students from class XI IPA 1 as conrtol group and class XI IPA 2 as experimental group. In collecting the data, test was used. The test was given twice to both experimental and control groups, as a pretest and posttest. To verify the hypotheses, the data of pretest and posttest were analyzed by using independent sample t-test and paired sample t-test in SPSS program. The findings showed that the p-output from paired sample t-test (sig2-tailed) was 0.003 which was lower than 0.05 and t-value 3.254 was higher than t-table with df=29 (2.0452). The result of p-output from independent sample t-test was 0.022 which was lower that 0.05 and the t-value 2.355 was higher than t-table with df=58 (2.0017). It means that teaching narrative reading by using Feature Walks strategy had significant effect on the students’ reading comprehension scores.