In an increasingly competitive hospitality landscape, customer loyalty remains a key driver of long-term success. While many studies have explored its antecedents, the findings are often fragmented due to differing methods, contexts and timeframes. This meta-analysis addresses that gap by synthesizing results from 169 empirical studies published between January 2000 and September 2024, offering a comprehensive and time-sensitive understanding of what drives loyalty in the hospitality industry. We distinguish between two periods: a phase of relative economic stability (2000–2019) and a period marked by multiple global crises (2020–2024). By applying and extending theoretical models, we identify the key variables that directly influence different types of loyalty. Effect sizes are calculated and compared and potential moderators, such as macroeconomic indicators (e.g., inflation, GDP) and cultural dimensions (e.g., power distance), are systematically examined. The findings show that in stable times, emotional connection plays a central role in fostering loyalty. In contrast, during periods of crisis, rational factors such as switching costs become more prominent. Both cultural and economic conditions significantly shape the impact of these drivers. Our study contributes to a deeper, more nuanced understanding of loyalty development under shifting global conditions. The results not only advance cross-cultural and longitudinal loyalty research, but also offer concrete guidance for designing adaptive, crisis-resilient loyalty strategies in the international hospitality sector. • Meta-analysis of 169 studies on loyalty drivers in the international hospitality industry (2000–2024). • Reveals time-contingent loyalty dynamics with economic and cultural moderators. • Finds emotional antecedents weakens post-2020; switching barriers gain predictive strength. • Identifies trust and satisfaction as less important under macroeconomic pressure. • Provides evidence-based guidance for context-specific loyalty strategy design under global uncertainty.
Reitenberger et al. (Fri,) studied this question.