Social distancing due to the COVID-19 pandemic has restricted socially supportive exchanges to a non-face-to-face method. This study examined whether and how both socially supportive exchanges (i.e., face-to-face and non-face-to-face) help the recovery process. Based on the conservation of resources theory and the complementarity hypothesis, we hypothesized a moderated mediation model where positive affect mediates the effects of both socially supportive exchanges on recovery, and the level of the face-to-face exchanges would moderate the mediation effect of the non-face-to-face exchanges. Multilevel linear modeling analyses of data collected from daily surveys of 118 university students over two weeks largely supported the hypothesized model. Specifically, both socially supportive exchanges positively affected the recovery outcomes (i.e., state of being recovered and sleep quality) through positive affect. Although the moderated mediation effect was not statistically significant, a simple effect analysis showed that the positive effect of non-face-to-face socially supportive exchanges on the recovery process existed only when the days of face-to-face socially supportive exchanges were low, which is consistent with the hypothesized model. The findings of this study shed light on the potential positive effects of non-face-to-face socially supportive exchanges, especially when face-to-face social interaction is restricted.