It is a common observation that operating companies are reluctant to use EOR techniques when they have the option of infill well drilling instead. Reasons for this include the advances in technology that allow accurate prediction of where unswept reserves are located. Other factors include the quicker recovery of the investment, and the fact that oil companies are well experienced in judging the risk and possible returns associated with drilling wells, whereas quantification of the risk and possible returns from EOR projects is difficult to calculate in advance and to evaluate once in place. However, are the oil companies missing opportunities for maximising return on their investment, and are there limitations in the way in which risk is evaluated which inherently favour infill well drilling? This study involved setting up a range of EOR scenarios, performing reservoir calculations to evaluate additional recovery and then making a comparison with infill drilling to maximise recovery from mature assets. This paper develops an understanding of how the economic comparison should be performed, since the timescales for investment and return on investment and the associated risks and uncertainties are different for EOR projects and infill well drilling. One driver is whether the oil price is relatively high or low. The method involves studying a range of scenarios, selecting appropriate EOR techniques and modelling the impact these techniques have on recovery, and then running calculations of the impact of various options. An economic assessment is made of the costs and risks of the various options together with expected return under a range of economic scenarios. For illustration, a synthetic reservoir simulation model is developed to study the impact of polymer flooding vs. waterflooding. Various uncertainties are tested, and the results are fed into an economic model for an evaluation of sensitivity to various parameters. This was done for polymer flooding, but the approach could be used for other EOR techniques also.