Setting aside parcels of land is the main conservation strategy to reduce the rate of biodiversity loss worldwide. Because funding for biological conservation is limited, it is important to distinguish the most efficient ways to use it. Here, we assess implications of alternative measures to conserve biodiversity in managed boreal forest landscapes. We calculated four alternative spatio-temporal scenarios and compared these to the current management regime over 100-year time period. In the alternative scenarios, a fixed amount of funding was invested in (1) permanent large reserves (each tens of ha in size), (2) permanent small reserves (each a few ha in size), (3) temporary small reserves (based on 10-year contracts with private land owners), and (4) green-tree retention (small groups of trees retained on clear-cuts). To assess biodiversity implications, we used habitat suitability indices to calculate overall habitat availability for five groups of red-listed and habitat-specific species associated with decaying spruce logs. The possibilities for timber harvests did not differ among the scenarios, but biodiversity performance was different. The scenarios with permanent reserves tended to outperform other scenarios, suggesting that conservation policies based on permanent reserves are the most cost-efficient in the long term. Results, however, varied among time scales and species groups. In the short term, a strategy of investment in temporary small reserves was the most efficient. Habitat for species associated with old spruce dead-wood and preferring shade was rare throughout all simulations, and therefore, it is likely that these species cannot be sustained in managed forests. Species that live on fresh deadwood and are associated with forest edges coped well in all scenarios suggesting that such species will persist in managed landscapes without additional conservation Communicated by M. Moog.