This study evaluated the utility of structured descriptive assessment (SDA) to generate a hypothesis regarding the operant function of problem behavior when the analogue functional analysis (FA) failed to evoke problem behavior for an adult with developmental disabilities. The effectiveness of interventions based on that hypothesis was evaluated in the natural environment. The SDA succeeded in producing a relatively controlled baseline of problem behavior where the FA and direct observation could not. However, the extent to which treatment procedures affected problem behavior could not be determined due to confounding variables outside the control of the experimenter. The results provide cautionary evidence highlighting both the potential utility of SDA and challenges that may be encountered when conducting SDA and evaluating treatments in natural environments.