Using text and image mining to study how actor engagement creates value in the sharing economy

Abstract

Purpose This paper aims to explore the concept of “actor engagement” within the context of the sharing economy, a novel and dynamic business model. Specifically, it investigates the formation of actor engagement and its relationship with value creation within this business model. Design/methodology/approach Drawing on Storbacka et al. (2016)actor engagement framework and service-dominant logic service ecosystem model, unstructured data (text and images) from the Airbnb platform in seven countries and text- and image-mining techniques such as machine learning are used to measure the research variables and test the model by PLS-SEM. Findings The results indicate that affective engagement has a more significant impact on behavioural engagement than cognitive engagement for multidimensional actor engagement. Service providers’ engagement – directly, and through customer engagement – influences value creation for service providers (i.e. performance). Moreover, national-level moderator (i.e. economic, competitiveness, technology, social and political factors) plays a significant moderating role in our model. Research limitations/implications This study encourages future research to explore how actor engagement leads to value creation for all actors on the different sharing economy platforms. Practical implications The findings provide practical insights for service providers to engage their customers and platform managers, especially in an international context, on managing their relationships with both customers and service providers in different countries. Originality/value This study advances the current literature on actor engagement and its role in value creation by providing a better understanding of the role of the national context in this process through unstructured data analysis.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 7 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.