While there are ways to integrate them in order to accountfor this heterogeneity, econometric estimates of a parametric kindare not fully equipped to accurately illustrate its impact on pro-duction and its relevance for innovation. However, less standardsemi-parametric techniques can be used to this scope and inter-esting implications can be obtained from them. This is particularlythe case for quantile regression, on which the present paper focuses,aiming to show that it can be a useful analytical tool for a micro-econometric estimate that directly tackles firms’ heterogeneity andthus increases its utility in the analysis of innovation. In partic-ular, the quantile estimation can help us detect how much theproduction impact of firms’ inputs varies along different firm-sizequantiles and in different economic sectors.