Decisions regarding the management of civil infrastructure are becoming more crucial as a large share of bridges is presently approaching what is often considered to be the end of their theoretical service duration. Evaluating existing structures using a data-driven approach rather than subjective visual inspections, conservative modeling assumptions and unrealistic re-calculations can avoid replacing existing infrastructure prematurely. Structural performance monitoring uses field measurements to provide more accurate evaluations of bridge behavior. As the goal of this monitoring method is to verify bridge safety at a given time, it should be differentiated from structural health monitoring, which aims at detecting structural damage. Possible techniques for structural performance monitoring include non-destructive testing, bridge load testing, and continuous monitoring of structural behavior, load levels, and environmental conditions. Nonetheless, selecting the optimal combination of monitoring techniques is challenging due to the difficulty in predicting their unique information gain and the redundancy in this information. Moreover, information collected on bridge parameters may have various influences on structural verifications, especially because different limit states are usually considered. The value of information must be evaluated before monitoring to ensure that collected data can impact engineering decisions regarding structural safety. This study proposes a method to assess the value of information from multiple bridge monitoring techniques. A riveted steel railway bridge from 1897 in Switzerland is taken as an example. The optimal monitoring technique is defined based on the effects of uncertainty reductions on structural verifications and monitoring costs. Field measurements collected through bridge load testing and continuous monitoring validate results in terms of value-of-information predictions.