The vast amount of available data, computational advances, and increasing social demands from customers present enormous challenges for delivery companies. Recognizing this need, we aim to investigate how such companies manage customer claims through X®. Our research, employing a sequential mixed-methods approach, began with a literature review through bibliometric analysis and subsequent interpretation via content analysis, concluding with the triangulation of theoretical findings with empirical evidence obtained from social media. For the final analysis phase, we collected and analyzed user mentions of delivery brands on X®, creating a data corpus-i.e., a sample collected through techniques-essential for our exploratory analysis. In this big data sample, we applied various natural language processing and machine learning algorithms, uncovering how users of these delivery companies tend to compare brands when making complaints. Furthermore, our research highlighted the psychological bias of users, who tend to polarize between love and hate for brands, respond to other user's posts, and engage in significant interactions with likes. Consequently, factors such as brand comparison, polarization between love and hate, and user interaction emerged as the main predictors of claims to these companies, underscoring the practical implications of our findings for the delivery industry. © 2024 (Escobar-Sierra, Guisao Giraldo).