One of the limiting factors to characterize the soil spatial variability is the need for a dense soil sampling, which prevents the mapping due to the high demand of time and costs. A technique that minimizes the number of samples needed is the use of maps that have prior information on the spatial variability of the soil, allowing the identification of representative sampling points in the field. Management Zones (MZs), a sub-area delineated in the field, where there is relative homogeneity in yield potential, due to similar soil nutrients and environmental effects caused by similar landscape or soil conditions, has been widely accepted in management systems willing to apply precision agriculture techniques. MZs can be delineated using soil apparent electrical conductivity and local relief conditions variability, based on the clustering algorithm. Once the MZs relatively similar in terms of soil characteristics are created, there are no longer the need to take many soil samples to characterize the field. However, remains the question: where the best places are to take a limited number of soil samples within the MZ, where the mean value obtained from these samples represents the overall mean value of the attribute corresponding to the entire zone. This study aimed to evaluate a methodology to define the best locations for soil sampling to represent with a proper resolution of the physic-chemical soil variability. The best location for each soil sample was defined using the Fuzzy C-Means algorithm with some modifications. To evaluate the performance of the propose methodology, one field of 93 ha of sugarcane, was used to delineate the management zones and select the best places for soil sampling. The results obtained using 1 sample per 3 hectares by a guided sample grid was compared with the regular soil sampling, which uses, approximately, 1.3 sample per hectare, evaluating the soil physic and chemical attributes. The results show that, with this approach, it is possible to create targeted sampling grids with high precision for specific nutrients management, reducing costs and increasing the sustainability.