Purpose Mosaicism is a prevalent characteristic of human preimplantation embryos. This retrospective cohort study aimed to investigate pregnancy outcomes after transfer of mosaic or euploid embryos. Methods The embryos, which had been transferred as “euploidy,” were processed using array-based comparative genomic hybridization (aCGH). The original aCGH charts of the transferred embryos were reanalyzed. Mosaic and control euploid embryos were defined according to log2 ratio calls. Results Overall, 102 embryos were determined to be mosaic, of which 101 were estimated to harbor no more than 50% aneuploid mosaicism. Additionally, 268 euploid embryos were matched as controls. The rates of live birth (46.6% vs. 59.1%, odds ratio (OR) 0.60, 95% confidence interval (CI) 0.38–0.95), and biochemical pregnancy (65.7% vs. 76.1%, OR 0.60, 95% CI 0.37–0.99) per transfer cycle were significantly lower after mosaic embryo transfer than after euploid embryo transfer. The rates of clinical pregnancy and pregnancy loss and the risks of obstetric outcomes did not differ significantly between the two groups. Conclusions Compared with euploid embryo transfer, mosaic embryo transfer is associated with a lower rate of live birth, which is mainly attributed to a decreased rate of conception. However, as mosaic embryo transfer yielded a live birth rate of 46.6%, patients without euploid embryos could be counseled regarding this alternative option.