This article provides a frame for evaluation of natural resource interventions, which necessarily involves both human and natural systems. Two-system evaluands require us to adapt evaluation methods for comparison and attribution and to address differences in time and space occurring across the systems as well as potentially very different values among stakeholders. While two-system evaluands can be challenging, it does not follow that evaluation in these settings is necessarily more difficult than evaluations located solely in human systems settings. This article suggests that if we are concerned about use and influence, then our responses to challenging evaluation settings should not automatically favor additional rigor; they should prioritize salience and legitimacy through joint knowledge production processes with decision makers and stakeholders. (Contains 3 notes.)