Abstract Relapse is an important clinical and research problem in the addictions, but one that has no consensus on an operational definition. How relapse is measured could make substantial differences in clinical practice and in conclusions drawn from empirical studies. The aim of this study was to compare and contrast four different operational definitions of alcohol relapse in a clinical sample of adolescents. The participants were 75 males and females ages 14–18 who presented for outpatient treatment of alcohol use disorder (AUD). They completed an extensive in-person initial assessment, as well as 12 subsequent monthly telephone interviews that concerned their daily alcohol and other drug use and related events. The four alcohol relapse definitions evaluated included: at least 1 day of drinking any amount after at least 4 consecutive days of abstinence; at least 1 heavy (five standard drinks for boys, four for girls) drinking day after 4 abstinent days; at least 1 day of drinking any amount with associated problems after 4 abstinent days; and at least 1 heavy drinking day with associated problems following 4 abstinent days. Relapse events were identified during the first 6 months following the initial interview. The results showed that the time period until relapse varied across the four definitions of relapse ranging from 26 to 90 days. Additional analyses investigated the utility of each of the definitions in predicting functioning during months 7–12 using multiple regression. The results of these analyses showed that the two “heavy drinking” definitions contributed significantly independent variance to predicting average number of drinking days/month and drinks/drinking day and that the two “any drinking” definitions significantly predicted the presence of a current AUD diagnosis. The results suggest that relapse definitions make a difference in estimates of rates and times to first relapse. Furthermore, the occurrence of at least one relapse is predictive of later functioning in a clinical sample of adolescents, which is of practical and theoretical importance. Suggestions for the extensions of this investigation in future research are provided.