Upper limb impairments are common in women following surgery for breast cancer. Range of movement (ROM) exercises are commonly prescribed, but the optimal timing to begin these exercises is not clear. The aim of this systematic review and meta-analysis was to evaluate the effectiveness of early ROM exercises (Early ROM) compared to delayed ROM exercises (Delayed ROM) or usual care (UC) in reducing common complications in women following breast cancer surgery. Cochrane Central Register of Controlled Trials, Pubmed, EMBASE, CINAHL, and PEDro were searched from date of inception until the 15th of February 2021, to identify randomized controlled trials which compared Early ROM to either Delayed ROM or UC in women following surgery for breast cancer. Outcomes included shoulder range, wound outcomes and lymphedema incidence. Risk of bias was evaluated with the PEDro scale. Data analysis was conducted in R (version 3.6.0), with a priori sensitivity analyses conducted for studies with low risk of bias and published after the year 2000. Of the 703 articles retrieved, 20 trials (2442 participants) were eligible for inclusion. There were few differences between groups in ROM, except for flexion ROM when more recent evidence was considered. Total drainage time and hematoma incidence were significantly higher in the Early ROM group compared to Delayed ROM (WMD 1·2 days [95%CI 0·7,1·7], and RR 1·6 [95%CI 1·1,2·3], respectively). When considering more recent evidence, there were no differences between groups for these outcomes. There were no differences between groups in other wound outcomes. Lymphedema incidence was higher in the Early ROM group compared to Delayed ROM in the short-term only (RR 3·7 [95% CI 1·3;10·9]), and there was no difference when compared to UC. The quality of evidence using the GRADE approach was generally low to very low. This review found that when considering more contemporary evidence, the timing of exercise may influence ROM, but not wound outcomes. Further research is recommended to understand the effect on lymphedema incidence. No funding was sought for this review. A protocol for this systematic review was posted on the Open Science Framework prior to commencement (DOI: 10.17605/OSF.IO/Q5FHS).