Complicated problem. Prospecting criteria depend largely on possible deposit models in the analysed region (which geological structure is known). Start with choosing proper deposit types/models which were described eg. in: Cox&Singer (1986) USGS Bull. 1693, Briskey&Schulz (2002) USGS Open File Rep. 423, Emsbo (2009) USGS Open File Rep. 1209, Leach&Taylor (2009) USGS Open File Rep. 1213, Shanks et al. (2012) USGS Sci.Invest. Rep. 5070-C etc. They are taught during relevant courses, eg. Ore Deposit Models and Exploration Strategies, Tasmania 2015, KEA701.
At a very early stage determine mediangrade and tonnage in each regionally possible model and probability of finding an ore deposit of sufficient quality/size to be economically exploited. Transform tonnage into expected size of ore bodies, inclination (inferred from geological structure) and depth. Build graphic model (cross sections) with expected luithologies and choose adequate geophysical methods sufficient to distinguish probable anomalies created by ore bodies in such geological environment.