Recent years witnessed a surge in research on early stage patterns in Internet dating, but family scholars still lack information about the social costs of finding a partner online. How the Internet becoming an increasingly prevalent partnership market contributes to the perpetuation of social class boundaries in union formation is yet to be clarified. More specifically, it is unclear which social group makes better use of the abundance of opportunities for interaction online to form committed long-term unions. Based on several theoretical arguments (e.g., related to social structure theory, the role of status preferences in mate selection), this study aims to understand if online tools for mate selection affect the marital prospects of the currently less prone to marry lower educated. By means of up-to-date German data spanning over several waves and a sample of N = 4,224 partnered respondents between 24 and 45 years old, I assess educational differences in probability of being part of a committed (i.e., cohabitation or marriage) versus a non-committed partnership among couples that met online in comparison to those that found their match elsewhere. Results based on multinomial logit modeling show that meeting online is associated with a reduced probability of being married for those with low educational levels. Supplementary analyses also show that highly educated singles looking for a partner online increase their partnering advantage by being even more likely to transition into a union when searching online. These findings have implications for accelerating the current trend of widening social inequalities in marriage.