Metformin is an effective alternative to insulin for nonpregnant women with type 2 diabetes mellitus; however, it has not yet been studied in the setting of gestational diabetes mellitus (GDM). Concerns have been raised about the effect of metformin on fetal development, particularly because it crosses the placenta. In this Practice Point commentary, I discuss the findings of an open-label, multicenter, prospective trial conducted by Rowan et al., in which women with GDM were randomly allocated to receive metformin (plus insulin when necessary) or insulin alone. The rate of a composite neonatal outcome and the efficiency of glycemic control were not significantly different between the groups. Compared with the insulin group, the prevalence of severe neonatal hypoglycemia was lower but the rate of preterm birth was higher in the metformin group. Of note, 46% of metformin-treated women required supplemental insulin. Although the results of this study are encouraging, further data are needed on the long-term safety of metformin before it can be considered as first-line therapy for women with GDM.