The aim of the present study is to investigate the extent to which (semi-)automated annotation systems for the analysis of spoken discourse in English as a Lingua Franca (ELF) in migration encounters can be considered as reliable for the pragmatic annotation of speech acts. The motivations which lie behind this are two-fold: on the one hand, the study was motivated by the need to provide a pragmatically-annotated corpus of spoken interactions which could be searched for by intercultural mediators for analysis and for their own professional training; on the other, a personal interest in the use of (semi-)automated research tool applied to the exploration of pragmatic aspects of discourse. The issue of whether automated software tools can actually capture pragmatic change is recurrent throughout the study and, as will be seen, represents the core of the analysis of ELF exchanges in migration encounters.