Solubility determination of organic molecules having very low solubilities is hampered by such problems as slow equilibration during measurement, influence of impurities, and inherent heterogeneity in the energetic content of the crystalline solid. Three approaches to meeting these problems are presented. The first approach involves enhancing the dissolution rate by the addition of a water-immiscible solvent in which the organic solute is more soluble, thereby increasing the surface area available for dissolution. The second method is a combination of experimental data with a group contribution approach that allows the estimation of extremely insoluble solids. This approach involves measurement of the solubility in an organic solvent and calculation of the aqueous solubility from the estimated partition coefficient and the organic solvent data. The third approach is based on using a large excess of the solid and a highly specific analytical determination of the main component. The first two approaches were explored in detail and tested using norethindrone, norethindrone acetate, methyltestosterone, and methyltestosterone acetate.