Interpreting interaction terms in econometric modelling is fussy. Even if logit or probit models are the most used modelling consumers' choice, they omit interaction effects among explanatory variables in the choice process. These blended effects are however declared in the modifications
of consumption decision. The difficulty is to interpret cofficients associated to these effects (interaction variables). To solve this problem, we propose a decision rule enforceable whatever the nature of the estimators. We build a convenient decision rule. We carry out an application of this decision rule to the choices of the wine consumers confronted with increasingly sophisticated products.