Statisticians frequently voice concern that their interactions with applied researchers start only after data have been collected. The same can be said for our experience with home-range studies. Too often, conversations about home range begin with questions concerning estimation methods, smoothing parameters, or the nature of autocorrelation. More productive efforts start by asking good (and interesting) research questions; once these questions are defined, it becomes possible to ask how various design and analysis strategies influence one's ability to answer these questions. With this process in mind, we address key sample-design and data-analysis issues related to the topic of home range. The impact of choosing a particular home-range estimator (e.g., minimum convex polygon, kernel density estimator, or local convex hull) will be question dependent, and for some problems other movement or use-based metrics (e.g., mean step lengths or time spent in particular areas) may be worthy of consideration. Thus, we argue the need for more question-driven and focused research and for clearly distinguishing the biological concept of an animal's home range from the statistical quantities one uses to investigate this concept. For comparative studies, it is important to standardize sampling regimes and estimation methods as much as possible, and to pay close attention to missing data issues. More attention should also be given to temporally changing space-use patterns, with biologically meaningful time periods (e.g., life-history stages) used to define sampling periods. Last, we argue the need for closer connections between theoretical and empirical researchers. Advances in ecological theory, and its application to natural resources management, will require carefully designed research studies to test theoretical predictions from more mechanistic modeling approaches.