Stream gages that measure river discharge are critical tools for monitoring and managing environmental flows. Yet stream gage infrastructure worldwide is underfunded, resulting in inadequate spatial coverage for critical management and scientific purposes. Previous approaches for identifying gaps in the gaging networks have failed to consider the geometry of interconnected streams. To allow stream managers to more efficiently and cost-effectively design and operate stream gaging systems, we present a method for evaluating the stream network coverage provided by an extant or hypothetical set of gages. We illustrate the approach in California, where we use information on drainage area and stream network structure to define current stream gage coverage. Next, we model scenarios for efficient gage network expansion and explore how a reconfiguration of gage locations could more efficiently address distinct management activities, including reservoir operations, biodiversity protection, and hydroclimatic monitoring. We determined that California’s stream network is poorly gaged, covering less than 10% of stream channel length, 10% of dams, and 30% of highly biodiverse watersheds. Reference-quality streams are also poorly gaged across the state, with limited representation of the state’s hydrologic diversity. Our analysis suggests that strategic installation of 500 additional gages on top of the more than 800 gages currently installed would increase the length of stream channel gaged by 15,748 km (from 11,288 km to 27,036 km), the count of dams gaged by 118 (from 75 to 193), the count of highly biodiverse watersheds by 338 (from 281 to 619), and the length of reference stream gaged by 11,102 km (from 287 km to 11,389 km). Results of this study provide a method for planning and evaluating stream network monitoring in the context of limited resources and multi-objective river management.