This paper proposes a construct of “behavioral planning” and presents a theory in which behavioral planning is an important cognitive antecedent to many health behaviors, particularly for those that have to be maintained over time. Behavioral planning is the cognitive elaboration of internal and external resources and constraints in the representation of action sequences. It is suggested that behavioral planning can vary in depth, duration, and frequency, and that this variation helps determine which beliefs become salient in the cognitive representations of behaviors. The relationships between behavioral planning and several of the cognitive constructs widely used in health communications — intention and self-efficacy — are described in detail. Two measures of behavioral planning and physical activity were included in a large-scale survey of the U.S. population. These items were normally distributed and stemmed from a single underlying factor. Analysis of the survey results indicated that this behavioral planning factor was related to measures of self-efficacy, intention, and self-reported behavior. These results lent support to the theory. A controlled experiment with random assignment to condition was conducted to test the causal directions of these relationships. It was hypothesized that behavioral planning would positively affect these variables. The effects of a manipulation of the depth and duration of behavioral planning were compared to both a control condition and a comparison condition, hi the latter, subjects were asked to think about the benefits of physical activity. The manipulations and instruments were delivered over the Internet using the World Wide Web. The results of the experiment did not confirm the hypotheses. Instead, there was some evidence for a counterproductive interaction effect between stage of change and behavioral planning. Among those subjects who were exercising somewhat regularly, the manipulation of behavioral planning had a negative effect on their moderate physical activity and no effect on their vigorous physical activity. There were several other statistically significant but unexpected findings. Speculating, there may be important differences between the frequency of behavioral planning and the depth and duration of planning.