Structures can be better understood when measurement data are used to improve the modeling of structural behavior. Our capacity to interpret data depends on aspects such as the choice of model class, model parameters (and their range of possible values), and the extent of uncertainties influencing models and measurements. The objective of this paper is to determine probabilistically to what degree measurements are useful for structural identification with respect to these aspects. A metric, expected identifiability, is proposed to be used prior to monitoring. The new methodology is based on three performance indexes: the expected number of candidate models, the expected prediction ranges, and a combination of the two. Because it does not require intervention on the structure, the method can be used as a tool to support prioritization of decisions related to full-scale testing. These features are illustrated through the study of the Langensand Bridge (Switzerland). In this example, the methodology shows that increases in modeling uncertainties significantly hinder the usefulness of measurements for identifying model parameter values. The predictive capability of the method proposed is verified by agreement with observations made during a recent structural identification exercise. Quantifying the expected identifiability provides a tool to support infrastructure decision making, such as determining to what extent certain structural monitoring plans are useful.