This work covers two separate field experiments. In the first one, the results of 1-ha grid soil analysis for soil organic matter (OM), pH, cation exchange capacity (CEC), nitrate N, P, K, S, Ca, Mg and soluble salts were compared with the results of yield mapping, biomass index from optical on-the-go sensors, as well as multispectral imagery analysis for the last 30 years.  As a result, it was found that none of the analyzed soil characteristics was predominant for determining yield. Correlation between the soil properties and yield of spring wheat was -0.24 for soil pH, -0.15 for phosphorus, -0.13 for nitrate nitrogen, -0.06 for potassium, 0.12 for soil OM, and 0.01 for soil electrical conductivity. At the same time, correlation between one-year normalize difference vegetation index (NDVI) and grain yield was 0.51, and multi-year NDVI resulted in r= 0.65. In the second experiment, we analyzed spatial variability of vegetation in the field using 22 layers of NDVI collected between 1984 and 2013, and compared these 22 years of data with one-year yield dataset to estimate the accuracy of management zones. Soil electrical conductivity (EC) measurement was also compared with the yield data. Correlation coefficients between one-year NDVI and yield data fluctuated between 0.3 and 0.75 depending on the year, and for soil EC the value of this coefficient was -0.34 for EC deep and -0.37 for EC shallow. Management zones delineated from soil EC data gave good separation for different soil types, but poorly separated areas with different yield potential. Field analysis of yield potential is contrast EC zones three weeks prior to harvesting revealed almost identical yield potential, whereas in the zones delineated from vegetation indices and yield data, the difference between high and low productive areas exceeded 200%. Based on our results, we concluded that yield is an integrated result of many different factors, including various soil characteristics, relief, PAR, air moisture etc., and it is very difficult to create an accurate model for yield planning based just on soil characteristics. The main goal of spatial analysis and delineation of management zones should be aimed to determine main yield limiting factors in the field. Yield data or vegetation indices obtained from multispectral satellite imagery give better results for delineation of management zones in the field than soil EC, and the accuracy of multi-year mapping is better than utilization of one-year data. Also, analysis of yield data or spatial variability of green biomass through various vegetation indices gives more predictable results for accurate yield goal planning than analysis of soil variability using grid sampling or soil EC measurement.