PurposeThe purpose of this article is to critically evaluate maintenance strategies and determine to what extent models from other sectors can be applied to building maintenance.Design/methodology/approachThe paper is a theoretical paper based on a number of Swedish studies, both case studies and questionnaires. From these a number of stylized facts have been identified and the purpose is to explain and draw conclusions from these.FindingsThe main finding is that there are a number of specific uncertainties that affect building maintenance planning making more detailed long‐term plans less meaningful. A new structure for maintenance is proposed focusing on long‐term strategies goals for various buildings/components and then short run adjustments when new information comes up.Research limitations/implicationsThe case for the new model needs to be strengthened by further studies, including studies from other countries.Practical implicationsMaintenance activities will be more important as the large building stock from the 1950s and 1960s needs to be renovated or demolished. A rational structure and realistic expectations concerning maintenance planning is then important.Originality/valueThe most important contribution of the article is to underline the importance of different types of uncertainty for the structure of maintenance planning for building.