In face-to-face communication people have problems recalling which facts they have disclosed to whom. We explore if such destination memory (DM) problems also exist in computer-mediated communication (CMC). Participants (N = 64) disclosed 50 pieces of personal information to 50 fictitious partners in a sham Skype environment in two conditions, one-sided “telling” and communicative “turn-taking”. Their recall for facts, faces, and fact-face-pairs was measured as dependent variables. ANOVA results show a significant main effect of type of recall (F (2,61) = 222.47, p < .001): recall for facts was better than for faces or fact-face-pairs, indicating that DM problems emerge when faces are previously unknown. Additionally, we found a significant main effect of conditions (F (1,62) = 6.75, p = .012): contrary to expectations recall in the “telling” condition was best. These interesting findings and those of a second study will be discussed considering the specific characteristics of CMC.