To reconcile the personalization-privacy paradox, we adopt the privacy-as-a-state view and define privacy as a state of information boundary rule-keeping. We further identify five boundary factors that underlie some of the important implicit rules to privacy keeping: communication channel, platform, device, time, and purpose. Using an online vignette survey, we investigate how each of the boundary conditions affected users’ privacy perceptions when they were subjected to personalized advertisements. With fixed and random effects models, we investigated how violating different boundaries leads to changes in perceived privacy. Our results show that all five boundary conditions are significant predictors of perceived privacy within individuals. Communication channel, device, and business versus private purpose are significant predictors of perceived privacy across the whole sample. Temporal boundaries and platform boundaries failed to achieve statistical significance when evaluated simultaneously with the other factors across the whole sample. This means that for each individual, observing these five boundary conditions will lead to higher perceived privacy than non-observant behavior; while taken as a whole, observing the communication channel, device, and business versus private purpose boundaries will lead to higher averages of perceived privacy in the whole sample. Theoretical and practical implications are discussed based on the results.