This paper describes the problems that arise in the translation of texts on – and by – the nonprofit sector from English (especially American English) into Italian. First of all, we focus on the main problems of translation, deriving from the disparity between the third sector in Italy and in the US. Secondly, we present a corpus-based project aimed at monitoring the differences between nonprofit discourse (especially fundraising letters) in Italy and the US. Lastly, we show a number of practical examples to illustrate how this type of corpus can meet the needs of translators and draw some conclusions on the importance of using monitor (i.e., continuously updated) corpora to produce successful translations in specialised fields where there are sizeable differences between source and the target cultures.