The state parameter is the main variable employed for predicting the undrained behaviour of artificial soils such as tailings and hydraulic fills. Current practice involves using screening methods and CPTu data (e.g. Robertson (2010)) or cavity expansion based methods (e.g. \nocite{Shuttle&Jefferies2016} Shuttle and Jefferies (2016)). However, these methods have drawbacks: they are based on empirical correlations for clean sands and do not consider the effect of partial drainage. This paper presents a site-specific procedure to determine the state parameter of tailings, inspired in the work by Monforte (2022). The procedure consists of: i) calibrating the CASM constitutive model using triaxial tests for different state parameters; ii) determining the plausible range of hydraulic conductivity based on dissipation tests; and iii) conducting numerical simulations of CPTu tests using the Pocket G-PFEM tool for different combinations of state parameters and hydraulic conductivities. By comparing the results of the simulations with the real CPTu data, a site-specific relationship between the state parameter and the CPTu measurements can be established. To validate the procedure, the method is applied to a real tailings deposit and contrasted against routine screening methods. In the particular case studied in this paper, the method predicts more contractive behaviour than the screening methods. While the method is still in an early stage of development, it looks very promising because it allows for using the raw CPTu data to calibrate a constitutive model, without resourcing to any kind of correlations or empirical transformation models.