This study examines to what extent travel information can be used to direct travelers to system-optimal routes that may be sub-optimal for them personally, but contribute to network efficiency. This is done by empirically examining determinants of travelers' compliance with social routing advice. To that end, we conducted both a stated choice experiment and a revealed choice experiment (which also collected stated intentions and motivations for revealed behavior). Results from the stated choice experiment indicate a significant difference in compliance behavior across different information frames, societal goals, sizes of travel time sacrifices and personality. These findings are less evident from results based on analysis of revealed choices; i.e., the main motivation for revealed compliance seems to be an intrinsic motivation to contribute to improved throughput, while the main motivation for non-compliance relates to perceived traffic conditions. Moreover, the size of the travel time sacrifice seems not that important as expected. Nonetheless, comparing stated intentions with real-world behavior suggests that a relation between intention and compliance frequency does exist.