This work aims at developing a two stage strategy which employ information on clustering schemes identified by a mapping analysis with the purpose of estimating a club convergence model. At the first stage, unobserved TFP differentials across regions are identified by introducing a mapping structure in a conditional convergence growth model. Since estimation of this class of convergence models in the presence of regional heterogeneity poses both identification and collinearity problems, we develop an entropy-based estimation procedure which simultaneously takes account of ill-posed and ill-conditioned inference problems. At the second step of the analysis, we estimate a two-club spatial convergence model, where clubs correspond to subsets of total observations, as identified at the first stage of the analysis and spatial dependence is modeled. The two step strategy is applied to assess the existence of conditional convergence across Italian regions over the period 1960-1999.