Background: The use of acellular hydrogels to repair osteochondral defects requires cells to first invade the biomaterial and then to deposit extracellular matrix for tissue regeneration. Due to the diverse physicochemical properties of engineered hydrogels, the specific properties that allow or even improve the behaviour of cells are not yet clear. The aim of this study was to investigate the influence of various physicochemical properties of hydrogels on cell migration and related tissue formation using in vitro, ex vivo and in vivo models. Methods: Three hydrogel platforms were used in the study: Gelatine methacryloyl (GelMA) (5% wt), norbornene hyaluronic acid (norHA) (2% wt) and tyramine functionalised hyaluronic acid (THA) (2.5% wt). GelMA was modified to vary the degree of functionalisation (DoF 50% and 80%), norHA was used with varied degradability via a matrix metalloproteinase (MMP) degradable crosslinker and THA was used with the addition of collagen fibrils. The migration of human mesenchymal stromal cells (hMSC) in hydrogels was studied in vitro using a 3D spheroid migration assay over 48h. In addition, chondrocyte migration within and around hydrogels was investigated in an ex vivo bovine cartilage ring model (three weeks). Finally, tissue repair within osteochondral defects was studied in a semi-orthotopic in vivo mouse model (six weeks). Results: A lower DoF of GelMA did not affect cell migration in vitro (p ​= ​0.390) and led to a higher migration score ex vivo (p ​< ​0.001). The introduction of a MMP degradable crosslinker in norHA hydrogels did not improve cell infiltration in vitro or in vivo. The addition of collagen to THA resulted in greater hMSC migration in vitro (p ​= ​0.031) and ex vivo (p ​< ​0.001). Hydrogels that exhibited more cell migration in vitro or ex vivo also showed more tissue formation in the osteochondral defects in vivo, except for the norHA group. Whereas norHA with a degradable crosslinker did not improve cell migration in vitro or ex vivo, it did significantly increase tissue formation in vivo compared to the non-degradable crosslinker (p ​< ​0.001). Conclusion: The modification of hydrogels by adapting DoF, use of a degradable crosslinker or including fibrillar collagen can control and improve cell migration and tissue formation for osteochondral defect repair. This study also emphasizes the importance of performing both in vitro and in vivo testing of biomaterials, as, depending on the material, the results might be affected by the model used. The translational potential of this article: This article highlights the potential of using acellular hydrogels to repair osteochondral defects, which are common injuries in orthopaedics. The study provides a deeper understanding of how to modify the properties of hydrogels to control cell migration and tissue formation for osteochondral defect repair. The results of this article also highlight that the choice of the used laboratory model can affect the outcome. Testing hydrogels in different models is thus advised for successful translation of laboratory results to the clinical application.