The purpose of this study was to improve state worker profiling models by: 1) establishing an approach for evaluating the accuracy of worker profiling models, 2) applying this approach to current state models to determine how effective they are at predicting Unemployment Insurance (UI) benefit exhaustion, and 3) identifying best practices in operating and maintaining worker profiling models. All 53 jurisdictions returned surveys while a smaller number of states (34) submitted claimant data. Basic assessments of model effectiveness were conducted for 28 of these states, and extended analyses were conducted for 9 of them. The extended analyses included updating the existing model with the data provided by the state, revising the model with transformed or derived data elements and the construction of an alternative model called a “Tobit” model. A full statistical treatment for the updated, revised, and Tobit models is presented. Research found that many states have not regularly updated the coefficients in their profiling models. In spite of this, the report shows the performance of profiling models to be reasonably good. Detailed analysis of state data shows that almost all of the 28 state models analyzed perform better than random assignment of claimants to services. The report also showed that virtually all models had better performance after recalculating coefficients or making relatively simple modeling changes.