Airbnb customers’ service quality and satisfaction with big data approach

A desire for sustainability, enjoyment of activities and financial gain, which has sparked a growing interest among researchers and businesses, drives the sharing economy phenomenon. The sharing economy enables people to sell services through reputable online platforms such as Uber or Airbnb. This research focuses on Airbnb, a peer-topeer (P2P) internet platform provider that has become one of the most successful models in the sharing economy. Because of the unique lodging experience that Airbnb users pursue, understanding Airbnb users’ perceptions of service quality and satisfaction by referring to the standards of traditional hotel customers can often be misleading. Therefore, more research is needed to explore Airbnb customers’ experience. Online reviews serve as the data source for this research, which provides a representative sample of individual customers’ personal and unique lodging experiences. This research is divided into two parts, Study 1 and Study 2. The primary purpose of Study 1 is to explore the Airbnb service quality attributes that are essential to deliver an outstanding customer experience. A novel structural topic model (STM) is employed due to its advantage that enables us to incorporate covariates in the analysis. Study 1 employed STM to extract service quality attributes from 242,020 Airbnb reviews in Malaysia. This study complements the lack of Airbnb-related research in developing countries and improves our understanding of the Airbnb service quality attributes in Malaysia. A widely used modified SERVQUAL questionnaire (MSQ) is cross-validated in this study by mapping identified service quality attributes to five service quality dimensions of this questionnaire, which contributes to the further modification of this instrument to suit the Airbnb context. By employing the methodological advantage of STM, this study extends previous studies by examining the different preferences of Malaysian and international Airbnb users. The results reconfirm the impact of nationality on customer preferences. In this study, Airbnb users from Malaysia are found to pay more attention to the property attributes (e.g., appearance, decoration); international Airbnb users are found to care more about whether this property is suitable for group accommodation, which could be associated with Airbnb users’ preferences for group travel. In addition, this study further examines the changing patterns of identified service quality attributes during a five-year period. The findings reveal the different changing patterns of Airbnb users’ perceptions of these attributes, notably, communication with the host and shopping are found to play an increasingly important role in Airbnb users’ experiences. The extracted service quality attributes perceived by Airbnb users provide a detailed reference for Airbnb practitioners to develop marketing strategies, property recommendation systems, and customer service standards. As for Study 2, it aims to investigate the drivers of satisfaction and dissatisfaction in the context of Airbnb accommodation, with a focus on Airbnb stay experiences.. The second study used LDA (Latent Dirichlet Allocation) and supervised LDA (sLDA) to achieve the study objectives, as these two topic models can effectively assist us in topic extraction and simultaneous analysis of quantitative data and topic distribution, respectively. A corpus that comprises 59,766 Airbnb reviews from 27,980 listings in 12 different cities is analyzed by using these two approaches. Unlike previous LDA based Airbnb studies, this study examines positive and negative Airbnb reviews separately. The results contribute to Airbnb literatures by revealing the heterogeneity of satisfaction and dissatisfaction attributes in Airbnb accommodation. In addition, the emergence of the topic “guest conflicts” in this study leads to a new direction in future sharing economy accommodation research, which is to study the interactions of different guests in a highly shared environment. The topic distribution analysis reveals the service attributes valued by users stay at different types of Airbnb properties, thus providing hosts operating different types of Airbnb properties with more targeted operational strategies to increase customer satisfaction. This study determines attributes that have the strongest predictive power to Airbnb users’ satisfaction and dissatisfaction through the sLDA analysis, which provides valuable managerial insights into priority setting when developing strategies to increase Airbnb customer satisfaction. Last, previous research highlighted the challenges of performing social media analysis, which is required to process an enormous amount of unstructured big data. Therefore, two studies in this thesis contribute to the development of the social media literature. Two user-generated content (UGC) analytical frameworks are developed for companies to use social media data in their service operation management. The detailed process of implementing these techniques is provided in this thesis and can serve as a useful reference for future research.

Text DING KAI - IR.pdf Download (1MB)

Actions (login required)

View Item