Currently the banks in Indonesia uses Selling Value of Tax Object (SVTO) present in Property Tax (mass valuation) valuating land collateral in association with loans they would provide to the customers. The purpose of this research is to produce a land valuation method, namely individual valuation based on environmental situation in certain land location. The data employed in this research include spatial data which uses Geographical Information System (GIS), such as searching for the nearest distance, searching for object in certain radius among others. Those data are then processed using multiple regression analysis (MRA). There are three main factors used in this study encompassing structural factor, neighbourhood factor and location factor. The research produced three models namely structural model, location model and spatial model. The results showed that structural factor contributed to land value variation at 56.5% and location factor contributed to land value at 36.7%. On other hand, the spatial model which combines structural and location model introduced to land value variation at 73%. The remaining 27% was influenced by non-spatial factors such as legality, economic situation and other factors. Test of residual had “Null Plot” form, suggesting that all asumptions for residuals which are acceptable to regression equation such as predictable linearity, constant error variance, independent errors and normal error distribution are satisfied.