In order to determine how sustainable online grocery shopping is as a practice, it is crucial to have an in-depth understanding of its drivers. This paper therefore validates the Unified Theory of Acceptance and Use of Technology (UTAUT2) in the context of e-grocery and enriches it with five constructs. We exploit a self-administered survey among 560 customers of two Belgian supermarkets and test the model by means of hierarchical multiple regression analysis. We do so not only for the full sample, but also for users and non-users separately. For the full sample, four of the five proposed context-specific constructs—namely, perceived risk, perceived time pressure, perceived in-store shopping enjoyment, and innovativeness—help better explain the intention to adopt or continue to use e-grocery services. In the subsamples, only perceived time pressure and innovativeness add explanatory power, and this only for non-users. In other words, the additional constructs primarily help discriminate between users and non-users. In addition, while the extended model outperforms the original UTAUT2 model for all three samples, the added value of the extended model does not so much lie in a higher explained variance, but rather in a more correct identification of the drivers of BI.