The present study evaluated a stepped-care model for the treatment of problem drinkers; those not severely dependent on alcohol. The initial treatment consisted of a motivationally based, four-session outpatient treatment. Based on previous research, treatment nonresponders were defined as having consumed more than 12 drinks per week between the assessment and third session. Six-month follow-up interviews were conducted on three groups of problem drinkers: (1) those who responded to the initial intervention (n=67); (2) those who did not respond to the initial treatment (n=36); and (3) those who did not respond to the initial treatment and received a supplemental intervention (n=33). The last two groups were used to evaluate whether providing treatment nonresponders with an additional “step” would improve treatment outcomes. The primary dependent measures were posttreatment percent days abstinent and posttreatment drinks per drinking day. Results suggested that: (1) within treatment drinking can help identify treatment nonresponse in stepped-care models; (2) the supplemental intervention did not influence posttreatment drinking; (3) treatment responders and nonresponders sought additional help at the same rate. The present study is the first study on stepped care for alcohol treatment and provides a methodology for evaluating stepped interventions. Recommendations for future research in this area include more attention to assessing the needs of treatment nonresponders and help seeking behavior of both responders and nonresponders after an initial intervention.