A novel metre-scale testing concept for studying geological carbon dioxide (CO2) storage (GCS) is presented, targeting a better understanding of the interplay between the different multi-physical processes and the time scales they occur. Unlike existing laboratory- and field-scale data, the proposed intermediate scale targets high-resolution measurements while accounting for spatial distribution. A metre-scale experimental setup is designed to reproduce stress and pressure conditions similar to the field. Synthetic caprock and reservoir geomaterials are used to simulate the geological GCS layout. A first compaction and water saturation phase of the entire system is foreseen before injecting carbon dioxide in the reservoir layer, after which, an extensive post-injection monitoring period follows until total hydromechanical equilibrium. Laboratory testing and numerical modelling of the hydromechanical behaviour of the caprock/reservoir system are used to design the different phases of the proposed campaign. The metre-scale experimental results will serve as a benchmark dataset to train data-driven machine learning algorithms for improving the performance of current numerical models and future predictions in large-scale GCS.