We examined videoconferencing in comparison with face-to-face office visits in primary care. Four physicians took part in the study and 175 patients were recruited from a general medical practice. Patients were interviewed and examined in both face-to-face and virtual settings, the order being randomized. Patients and clinicians were surveyed by questionnaire after each visit. Physicians were very satisfied with videoconferencing but preferred face-to-face overall (P < 0.0001). For videoconferencing, the physical examination and the ability to order appropriate laboratory tests were the least satisfying elements of the encounter. Patients were also very satisfied with videoconferencing but overall preferred face-to-face (P < 0.0001). This difference was significant for gastrointestinal, musculoskeletal and respiratory complaints. However, in terms of willingness to pay (WTP) for videoconferencing access, patients with musculoskeletal and respiratory complaints were the most willing. Age and gender did not significantly predict WTP. The technical quality of the videoconference had a significant effect on satisfaction with the clinical encounter but did not correlate with the patients' WTP for videoconferencing. Travel costs up to $40 and travel times up to 4 hours did not influence the WTP of patients. For non-rural, relatively mobile patients, videoconferencing appears suitable for short visits for relatively simple complaints.