Exploring Segmentation in eTourism: Clustering User Characteristics in Hotel Booking Situations Using k-Means

Stefan Eibl, Andreas Auinger, Robert A. Fina

Research output: Contribution to conferencePaperpeer-review

Abstract

In the dynamic field of eTourism, personalization and user segmenta-tion are paramount for enhancing user experience and driving digital platform success. This paper addresses the gap in eTourism research related to understand-ing consumer behavior through an external lens, due to the limited access to pro-prietary data from Online Travel Agencies (OTAs). We employ Adaptive Choice-Based Conjoint (ACBC) analysis and k-means clustering on data from a survey (n=801) based on 346 hotel listings on Booking.com, focusing on Vienna. Attributes such as star category, price, review valence, volume of reviews, scar-city indicators, sustainability cues, and city center proximity were examined to identify consumer preferences. Five distinct consumer clusters were revealed: Cost-Conscious Eco-Bookers, Green-Urban Deal Hunters, Social-Proof Assur-ance Seekers, Budget-Only Focused Minimalists, and Luxury-Quality Connois-seurs. These clusters vary in their prioritization of hotel attributes and de-mographics, demonstrating the diverse decision-making criteria within the eTourism market. This paper proposes a foundation for classifying user groups on booking platforms, enabling OTAs and hoteliers to tailor offerings to nuanced consumer segments, thus improving user experiences and potentially increasing conversion rates. The findings offer actionable insights into OTA personalization strategies and contribute to the scientific understanding of consumer behavior in the digital tourism landscape.
Original languageEnglish
Pages157-175
Number of pages19
DOIs
Publication statusPublished - 30 Jun 2024

Keywords

  • Personalization Tactics
  • eTourism Segmentation
  • k-Means Clustering

Fingerprint

Dive into the research topics of 'Exploring Segmentation in eTourism: Clustering User Characteristics in Hotel Booking Situations Using k-Means'. Together they form a unique fingerprint.

Cite this