When maintaining and restoring habitat, an important question is whether the spatial distribution of habitat affects its contribution to biodiversity conservation. The most straightforward way to test this is to assess the colonisation and extinction rate of habitat patches with different spatial locations in relation to potential dispersal sources. We did so, testing hypotheses developed within the framework of metapopulation theory. As a study species we used a red-listed beetle, Peltis grossa , which in our study landscape occurs mainly in high-cut stumps of a type routinely created at forestry operations in Sweden. At a per stump level, we analysed colonisation and extinction rates, and presence/absence in relation with high-cut stump succession, quality and connectivity. Most results were consistent with metapopulation theory, including a positive relationship between the colonization rate and a connectivity measure considering the species abundance in the surrounding area. However, the relationship became absent when using a connectivity measure based only on the occurrence of high-cut stumps in the surroundings. Thus, the spatial distribution of conservation efforts is important for this species: restored dead wood habitats are used by this species to a higher extent if situated close to large dispersal sources rather than further away. However, when detailed knowledge about dispersal sources is lacking, as it often is, connectivity is difficult to assess and include in conservation strategies.