The typology was derived from four variables (alcohol dependence severity, psychiatric severity, bad-driving index, and social instability) selected to maximize the feasibility of performing treatment matching with drunk- driving offenders. Five clusters that suggested specific treatment matching opportunities were identified. The largest, Cluster 4 (31 percent of cases), showed a low- problem profile; however, a moderate-severity group (Cluster 1), a high-risk driver group (Cluster 2), and two high problem-severity groups (Clusters 3 and 5) were also found. Clusters 3 and 5 had high levels of alcohol dependence and psychiatric symptoms but differed significantly on social instability. A low-intensity, alcohol-focused intervention would be appropriate for individuals in Cluster 1 because their dependence is only one criterion above the cutoff for diagnosis, and psychiatric severity and social instability are low. The treatment needs of those in Cluster 2 are likely to be best met with a moderate-intensity, alcohol- focused intervention and a program designed to enhance their driving skills. High problem severity in all areas indicates that Cluster 3 members need intensive treatment. Cluster 4 is likely to benefit from the lowest intensity intervention; and a high-intensity, broad-spectrum program that provides attention to psychiatric problems is indicated for Cluster 5. 3 tables and 60 references