Multi-level differentiation of short-term rental properties: A deep learning-based analysis of aesthetic design

This study aims to test the effects of differentiation on short-term rental performance along the dimension of aesthetic design. Online platforms display listing cover photos as search results, thus making aesthetic design a key element of differentiation. We hypothesize opposite impacts in two geographical scopes, local-and city-level, which answers an important question in differentiation literature of whom to compare to. Based on the assumption that localized competition has asymmetric influences, we introduce competition intensity as moderator. Hypotheses are tested with 96,196 listings from April 2021 to March 2022 in the Texas Airbnb market. We quantify aesthetic design by probability distribution scores over four design styles predicted by a pre trained machine learning model. This study identifies differentiation benefits at local-level but discounts at city level. Furthermore, it shows market intensity strengthens benefits and mitigates discounts regardless of the geographic scope. Finally, implications for aesthetic design as a strategic tool are discussed.

Automated summary

This study tests the effects of differentiation on short-term rental performance, finding that aesthetic design brings benefits at the local level but not at the city level. Additionally, market intensity strengthens the benefits of differentiation and mitigates the discounts.