Unlike for other asset classes the performance of direct property has been researched only to a small extent, especially as single-property data are hardly available. This study for the first time investigates the performance of direct property in Germany based on single-property data. Aims of the research are the identification of relevant market and property related performance factors, which lead to conclusions about the optimal allocation and selection of property investments. Due to the known heterogeneity of property it is obvious that this can best be achieved through investigating single-property data, as in this study. Cross-sectional analyses include indicators for market attraction and for the relative market positioning of each property. Among these are socio-economic data, property market data and geo-coded single-property data. Performance data and tenancy data were obtained from IPD. The investigated sample comprises 662 office properties, 184 retail properties and 585 residential properties, all of which were geo-coded and equipped with distance and driving times to local POI. Regression analyses were conducted with Total Return, Net Income Return and Capital Growth as independent variables. The results show that differences in the performance to a large extent depend on property management abilities. Micro-location particularly affects the performance of retail properties while socio-economic macro-location factors especially matter to residential properties. The valuation yield, as expected, is of major importance for property performance. Capital Growth and Net Income Return are largely affected by a different set of factors, while deviations between Total Returns mainly stem from the Capital Growth component.