This study examines a common pedagogical approach whereby students view teacher feedback outside of class time. Facilitated through Moodle, it aimed to investigate not only the effect of written corrective feedback (WCF) on learning outcomes, but importantly, how frequently students review the feedback they receive. The study was conducted at a Japanese Women’s University in Tokyo (age: M = 19.6, SD = 0.6). Over a 15-week semester, students were required to write weekly homework assignments. These were then marked by the researcher and returned via Moodle, with students encouraged to read their feedback before attempting the next assignment. In line with existing research in the field, a no-WCF control group was compared against treatment groups that received indirect WCF with meta-linguistic explanations. Further to this, using an online system such as Moodle allowed the provision of linked grammar resource pages through basic HTML code embedded within the WCF itself. At the end of the semester student texts were analysed to measure improvement. When pre-test and post-tests were compared there were no significant differences within groups or between groups on measures of fluency and accuracy. However, what sets this study apart from others that have investigated student engagement, is the use of online activity logs which in combination with survey data allowed the investigation of actual study habits rather than merely those that are self-reported. The results of which give better insight and valid explanations for the outcomes of the study and call into question the efficacy of take-home feedback.