Traditional museum audio guides have gradually been enhanced with multiple additional functionalities. The implementation of location-based services in particular, has enabled the collection of large volumes of spatiotemporal visitor movement data, from which individual visitor trajectories can be extracted. These trajectories can be studied to enable museums to better " know " their visitors, by means of trajectory mining. In addition, tapping into context data could make such analyses more expressive and revealing, by adding meaning and intention to trajectories. This work discusses important challenges in studying museum visitor movements with the help of context-aware indoor trajectory modeling and mining, and proposes work directions for each challenge.