Catchments that are almost ungauged, i.e. in which only a small number of point flow measurements are available, lie between the cases of ungauged and gauged catchments. Therefore it can be expected that blending model parameter estimation methodologies adapted to each of these two opposite cases may answer the problem of parameter estimation on almost ungauged catchments. The proposed method combines regional information (parameter sets of neighboring gauged stations) and local information (brought by the point measurements) in a framework where the relative weight of each source of information is made dependent on the number of point measurements available. This approach is tested with two different hydrological models on a set of 609 catchments. The results show that even a single measurement can improve the simulation’s efficiency, while ten measurements reduce more than 50% of the performance gap between the gauged and ungauged situations. The issue of possible information redundancy will be discussed.