This paper discusses the long-term mechanical behaviour of timber-to-concrete joints made with dowel-type fasteners. Despite the influence that the long-term behaviour of joints has on the mechanical behaviour of a timber–concrete structure and consequently on its design, there is still a lack of research in this area. This paper presents experimental research, carried out at the University of Coimbra and Delft University of Technology, on seven joint configurations using different types of fasteners and different materials. For each joint configuration, either four or ten tests were performed resulting in a total of forty tests. A comprehensive description of the test specimens and test setup is given. The experimental creep–time curves were fitted to a creep–time model and used to predict joint creep values over longer timeframes (10 and 50 years). The values obtained were compared with values available in the literature for timber-to-concrete joints with other types of fasteners and timber-to-timber joints with dowel-type fasteners. The approach for timber-to-timber joints suggested by Eurocode 5 was used to determine creep values for timber-to-concrete joints. The results obtained were compared with test results to assess the accuracy of predicting creep values of timber-to-concrete joints with dowel-type fasteners. It was concluded that creep values measured in long-term experimental tests are usually higher than those obtained from the model indicated in Eurocode 5, particularly for environmental conditions corresponding the service class 2.