Although many researchers have explored the use of Peer Feedback (PF) in writing (e.g., Hu & Lam, 2010), several have reported concerns with this technique, such as a tendency to shift most of the attention to micro features (e.g., mechanics, vocabulary) while giving little attention to macro features such as organisation and coherence (e.g., Van Steendam et al., 2010), even though macro features can be argued to be a highly important aspect of good writing (Truscott, 1996). This is one of the factors that have led researchers (e.g., Gielen et al., 2010b) to propose forms of this technique in which emphasis is placed on particular aspects of the PF process. This study introduces one such form of PF technique which requires learners to focus on macro features in writing and the teacher to focus on micro features, in order to give learners more time to critique essays at a macro level while receiving micro level FB from a reliable source. The study investigates the impact of the introduced form on: learners’ motivation to use PF and to learn writing; learners’ attitudes towards PF and towards writing; learners’ linguistic progress, and learners’ preference for giving and receiving macro and/or micro level feedback when practising PF technique. The research was conducted on 41 Saudi Arabian undergraduate students in their final year of an English degree course. An action research approach was adopted using a one-group design, with the PF activities divided into two consecutive phases. During the first phase, subjects practised the conventional use (i.e, providing PF on macro and micro features) of this technique (T1), while during the second phase they practised the new form of the technique (T2). The data were gathered over 15 weeks from pre-, mid- and post-tests; pre-, mid- and post-questionnaires; mid- and post-interviews; field notes; documentary evidence, and recording of several verbal protocol sessions. The findings suggest that both treatments can have a significant impact on the overall quality of learners’ writing, with the second treatment resulting in significantly better quality. Despite these findings, the learners showed a strong preference for conventional PF, suggesting they have difficulty in accepting the prohibition from providing PF on micro features owing to a negative transfer effect from their previous experience of approaches to teaching writing, which placed a great emphasis on the importance of micro features. It is likely that this transfer effect may be found in other contexts with a similar approach to teaching writing; further research is needed in order to test this hypothesis. In addition, in this study, the participants did not have the chance to see how much better they performed in their post-test, which raises the question of whether or not their views would have changed if they had.