Discrete trial training (DTT) is a well-established intervention for teaching skills to children with autism; however, few published guidelines are available for determining whether a child's rate of learning a particular skill is satisfactory and, if not, what to do. We assert that progress within 8-10 teaching sessions usually is evidence of satisfactory skill acquisition, whereas absence of progress within this time frame indicates a need to consider modifying or stopping instruction of that skill. Absence of progress may involve (a) consistently low rates of correct responding, (b) variable performance across sessions, (c) increases in problem behavior (often in conjunction with low or variable rates of correct responding), or (d) limited generalization of the skill outside intervention. Likely reasons for each of these patterns are described, and decision flowcharts for identifying possible solutions are outlined. When implemented with supervision from a qualified professional, these flowcharts may facilitate systematic problem solving. (Contains 2 tables and 6 figures.)