In comparative or matching research involving two or more treatments, the equivalence of the patient groups is of critical importance. In the past, equivalence has either been imposed by matching or balancing, or has been assured statistically by randomization. Matching and balancing, while useful in many contexts, nonetheless have important limitations, as does simple randomization. In recent years, a new tool has been developed that represents a compromise between balancing and randomization. This method, urn randomization, gives clinical investigators new options for improving the credibility of studies at a relatively modest cost. Urn randomization is randomization that is systematically based in favor of balancing. It can be used with several covariates, both marginally and jointly, producing optimal multivariate equivalence of treatment groups for large sample sizes. It preserves randomization as the primary basis for assignment to treatment and is less susceptible to experimenter bias or manipulation of the allocation process by staff than is balancing. Disadvantages include the fact that it is more difficult to implement, and that it violates the simple probability model of simple randomization. A number of research studies on addictions, including client-treatment matching trials, have used urn randomization. A summary of the mechanics of urn randomization is presented, and guidelines for its use in treatment studies are discussed.