Market research has developed several tools for selecting the optimal combination of product features. One of the methods which has become fairly popular within the last years is the "Trade-Off Model". With this method, utility indices are computed based on the evaluation of different product features through a sample of consumers. In our paper we are criticising the "Trade-Off Model" for not adequately describing consumer behaviour. We suggest using another product test approach, the "Experimental Pair Comparison Test", which permits an unbiased prediction of consumer behaviour. Empirical findings are presented to demonstrate the usefulness of the "Experimental Pair Comparison Test".