A full geotechnical risk analysis includes site investigation, model simplification and analysis. The site characterization forms the basis for the choice of the soil parameters used in the subsequent analysis, but the soil data is normally analyzed with crude methods that do not include the effect of the soil variability. A case study is here presented of soil characterization for later analysis of the stability of a light quay structure constructed on very soft Scandinavian clay. The field data consists of undrained shear strength measurements from field vane tests (FVT) and cone penetration tests (CPT), carried out at different locations along the structure. The measurements show a relatively large variation of the undrained shear strength, making a statistical characterization of the soil properties helpful for a later analysis of the reliability of the structure. In the current paper, the field measurement data (i.e. the tip resistance from CPT tests) are analyzed to obtain the correlation distance of the soil strength. The two types of measurements are combined to get an average of the undrained shear strength, while the spatial statistics are obtained from CPT measurements.