How do consumers form their reasons for/against chatbots? Symmetric variable relationship and asymmetric configurational models

Purpose Drawing on behavioral reasoning theory, this study aims to propose a research framework that investigates complex antecedent algorithms of reasons for (warmth and competence) and reasons against (realistic, identity and privacy threats) using chatbots in hospitality and tourism industries (H&TI). It empirically explores the breadth of usage (i.e. full vs partial usage) and demographics of travelers as antecedents of those reasons. In addition, the study examines symmetric directional theory and empirical analysis of how chatbot experiences relate to chatbot assessments. Design/methodology/approach Data were collected from 312 individuals in France. Data analysis was undertaken using a multi-analytical approach including symmetric variable analyses (PLS-SEM) and asymmetrical algorithm analyses (fuzzy set comparative qualitative analysis [fsQCA]). Findings The symmetric analyses did not validate all causal paths, whereas fsQCA evidences combinatory effects. This study finds that most full chatbot users judge chatbots as high in both warmth and competence. The study indicates that partial and full users evaluate chatbots differently, primarily with respect to their underlying reasons for usage. Demographics modestly shape travelers’ chatbot assessments. Combinations of demographics and use recipes are not accurate in indicating travelers assessing all threats in chatbot conversations. This last result can be because of a denial or coping mechanism that full users may develop to minimize the effects on these threats. Practical implications This study provides practical guidance for H&TI by proposing actionable recommendations to maximize reasons for (warmth and competence) and minimize reasons against (realistic, identity and privacy threats) using chatbots. Recommendations include, for example, providing clear information regarding chatbots’ real capabilities (i.e. its competence) and designing friendly interfaces to maximize perceived warmth. To counter threats, firms should strengthen their safeguards and promote these actions to users. Also employing big data analytics to segment users – full and partial users – and improve full users experience and convert partial users to full users would be beneficial. Originality/value To the best of the authors’ knowledge, this study is the first to show how recipes of reasons for (facilitating factors) and reasons against (hindering factors) using chatbots in H&TI are formed.