Measures of workers' inhalable dust exposures, tasks, and ventilation use were made in five small woodworking shops prior to the start of an intervention effectiveness study aimed at lowering personal wood dust exposures. The data were used to (1) design a sampling protocol for an intervention success measure, (2) identify targets for intervention among the tasks and activities responsible for high dust levels, and (3) develop shop-level measures as tools for tailoring intervention activities. Geometric mean dust concentrations ranged from 1.6 to 9.9 mg/m3 in the five shops, with the highest levels occurring in a cabinet shop. All shops had centralized dust collection systems and workers generally used dust control on stationary tools (60-100% of the time) when it was available. Sanding with both stationary and handheld powered tools, cleaning with methods that can disperse dust (e.g., brushes, compressed air), and miscellaneous tasks were all responsible for significant personal exposures. The positive association between miscellaneous tasks and exposures probably reflects the high background levels generated by nearby processes. Sanding with both stationary tools and handheld powered tools represents the most significant influence on personal exposures in small woodworking shops. The authors conclude that pilot studies are useful tools for designing occupational health and safety intervention effectiveness studies.