The most interesting part of this is their use of user data. Traditionally, it seems like there isn't a great way to judge demand for a restaurant until you open it, so it makes sense that user tracking (to gauge interest) and the ability to operate out of an already profitable space (reduced upfront cost) gives establishments a significant advantage (although the article doesn't exactly comment on number of failed online restaurants, so selection bias cannot be ruled out.) I wonder how competitors like GrubHub use data to gauge interest...