OBJECTIVE This study evaluated the accuracy of electroneurography to predict the prognosis of Bell's palsy and Ramsay-Hunt's syndrome. METHODS A retrospective, institutional review board-approved study. METHODS A secondary referral and a university-based centre. METHODS The patients had been treated for a sudden onset unilateral facial paralysis over the past 10 years (1994-2004). This retrospective study included only those patients who had been followed up for at least 3 months or if they had reached a complete recovery before then. METHODS House-Backmann grade versus electroneurography value. RESULTS The recovery rates to House-Brackmann grade II or better were 95% in those with Bell's palsy and 84% in those with herpes zoster oticus. The electroneurography value of the recovery and non-recovery groups from those with either Bell's palsy or herpes zoster oticus was similar. The logistic regression model between the electroneurography values and the probability of recovery showed no correlation in those with Bell's palsy or with herpes zoster oticus. This study did not identify the proper electroneurography value that had enough appropriate sensitivity and specificity to predict the prognosis of paralysis accurately in Bell's palsy or in herpes zoster oticus patients. CONCLUSIONS Electroneurography performed between day 7 and 10 for Bell's palsy or day 10 and 14 for herpes zoster oticus does not provide accurate information on the prognosis or recovery rate of the facial paralysis.