Abstract
Peer-to-peer accommodations have transformed the urban tourism sector, giving travelers flexibility and affordability due to Airbnb's popularity in this accommodation space. Airbnb has changed the norms of what a host property can be, far from just a simple lodging, as its brand slogan "Belong Anywhere" seeks to create authentic experiences and positions accommodation as an experiential product related to local culture and a sense of belonging. However, the increase in listings (over 8 million) has led to serious competition between hosts and commodification, making it increasingly difficult to continually meet guests' varying expectations.
The Experience Economy presents a compelling solution for differentiation by providing engaging experiences in four domains: education, entertainment, escapism, and esthetics. Previous studies have examined the four realms in bed-and-breakfasts, festivals, and cruise tourism, confirming its applicability in tourism, hospitality, and that experience varies depending on context. In Airbnb research, there is a recognition of the experiential aspects of the stay (home-feeling, interaction with hosts, and local embeddedness). While these studies confirm the validity of Pine and Gilmore’s framework across different hospitality settings, their application to peer-to-peer accommodation remains limited.
Existing research has largely relied on conceptual frameworks or small-scale survey data, leaving a gap in understanding how these experiential dimensions are reflected online. The studies underscore the significance of reviews, but have not applied the framework to Airbnb as a review dataset broadly, thereby restricting generalizability.
Comparative analyses across destinations are even rarer, despite the evidence that Airbnb reviews properties differ by market. Addressing these gaps is crucial to linking the Experience Economy framework with Airbnb context, where guest experiences are co-created and publicly expressed through online reviews.
Two cities, Paris and Rome, were selected for their large and mature Airbnb markets. Paris, a global hub of art and esthetics, exemplifies the modern tourist spectacle and a site of cultural exchange. Rome, rich in heritage assets, is shifting from a “museum city” to a dynamic hub of cultural events and urban competition.
Data consists of Airbnb guest reviews from Rome and Paris (approximately 1 million reviews for each city). Natural language processing has been increasingly used to analyze user- generated content in tourism and hospitality, enabling large-scale extraction of experiential attributes from reviews. The pipeline involves seed terms, human coding for validation (N ≈ 200), and the use of a large language model, involving Python, CSV agents, and the LangChainOpenAI coding environment. Guest satisfaction serves as dependent variable, measured through review ratings and sentiment analysis to capture objective and subjective satisfaction. Multilevel regression models will test the relationship between experiential dimensions and satisfaction, accounting for listing and city-level effects.