The impact of linguistic choices and (para-)linguistic markers on the perception of Twitter complaints by other customers: an experimental approach

Abstract This paper addresses how the realizations of different constitutive components of Twitter complaints shape the perception of these complaints by other customers. We present three experiments on French language in which we test how customer complaint perception is impacted by the realization of the complainable (Exp. 1), of the entity responsible for the complainable (Exp. 2), and of the customer’s wish for the complainable to be remedied (Exp. 3). The results of Exp. 1 indicate that the perceived likelihood that the complaint will be responded to by the company is highest when the complainable is realized as a combination of an assertion + question + picture. In Exp. 2, we found that, in comparison with the use of the discourse marker dites to refer to the entity responsible for the complainable, the use of a noun phrase or the absence of this component increases perceived politeness. Finally, our data from Exp. 3 reveal that, compared to the use of an imperative to voice the customer’s wish for the complainable to be remedied, “indirect” request forms, and preparatory interrogatives, in particular, are perceived as more polite, as expressing lower dissatisfaction, and as decreasing the likelihood of a response from the company.