With a considerable emphasis on the role of feedback in L2 writing, the effectiveness of written corrective feedback (WCF) has been investigated extensively over the past 25 years. Conflicting findings have been reported regarding the efficacy of WCF in developing learners' written accuracy. This paper provides an in-depth analysis of research on three key variables impacting the effects of WCF: feedback treatments; writing tasks with more/less cognitive and reasoning demands; and L2 written accuracy measures. Forty-two primary studies published between 2000 and 2020 were retrieved and coded following PRISMA guidelines. Results revealed variations in both feedback treatments and written accuracy measures with distinctive advantages and pitfalls. Further divergent issues concern the application of different genres of writing tasks that demand learners' various cognitive and linguistic efforts. These variations make it difficult to compare results across empirical studies. The findings contribute to a better understanding of why and how uniform criteria for selection of writing tasks and accuracy measures can ensure the comparability of studies.