Aim: To highlight and examine apparent paradoxes in assessing the effectiveness of different forms of land-use for biodiversity conservation. Location: Tanzania. Methods: We compare and contrast the findings of two recent and seemingly conflicting studies on the effectiveness of conservation protection strategies in Tanzania. We evaluate these studies in the context of a wider body of evidence relating to the problem of determining protected area performance. Results: We highlight the importance of landscape-scale management approaches for biodiversity conservation; establishing clear management and monitoring objectives in advance; the interrelation between the choice of target species and the appropriate spatial scale over which to measure their fate; and differences between snapshot and longitudinal scales in assessing the effectiveness of conservation strategies. Main conclusions: Protected area assessments should not promote an isolated focus on particular conservation targets or methods of monitoring and evaluating the effectiveness of conservation strategies. Instead we argue for a more pluralistic approach to evaluating conservation performance that can help to reveal where potential synergies in tackling different objectives exist, and clarifying the trade-offs when they do not.