In this work we explore the interplay between entry, selection and innovative learning as determinants of industrial evolution. We propose a model aimed to capture the essential features of learning and competition as drivers of the dynamics. Using both analytical and numerical techniques, we are able to disentangle possible generic properties which robustly hold for a wide range of parameterization. In particular, we identify different generic ``evolutionary archetypes'' in turn defined by characteristic interactions between entry/exit regimes, learning and industrial structures.