Rural e-commerce in China, particularly the rise of "Taobao villages" and "Taobao towns" (e-commerce hubs on the Taobao platform), plays a significant role in rural revitalization. Existing research focuses primarily on Taobao villages, neglecting the spatiotemporal dynamics of Taobao towns. This study aims to address this gap by examining the spatial distribution, temporal evolution, and influencing factors of Taobao towns across China. The study's importance stems from the potential of Taobao towns as a scalable model for rural development and transformation, offering valuable insights for other developing countries. The research questions address the growth stages of the transition from Taobao villages to Taobao towns, the spatial characteristics of Taobao town distribution, and the main factors influencing their development and mechanism of action.

Existing research on Taobao villages often focuses on case studies analyzing formation mechanisms and evolution processes, exploring the dynamic mechanisms of rural e-commerce, and identifying replicable practices. Studies highlight the roles of government support, talent development, logistics networks, and infrastructure upgrades. Research also examines the impact of e-commerce concentration on various aspects of rural life, including industrial structure, income, spatial form, innovation, and social environment. The spatial distribution of Taobao villages is generally characterized by a decreasing gradient from coastal to inland areas, but the central and western regions face challenges in sustained growth. In contrast, research on Taobao towns is limited, often treated as a supplementary perspective to Taobao villages. This study bridges this gap by providing a comprehensive analysis of Taobao town evolution using a spatial perspective.

This study employs a multi-method approach combining spatial analysis techniques to analyze the growth stages, distribution characteristics, and influencing factors of Taobao towns. Global and local spatial autocorrelation models (Global Moran's I and General G index) are used to assess the spatial association and clustering patterns of Taobao towns. Standard deviation ellipse analysis is employed to characterize the spatial distribution's central location, evolutionary direction, and dispersion trend. Finally, the GeoDetector model is used to quantify the influence of various factors on the spatial distribution of Taobao towns, including GDP, cargo volume, express business volume, road network density, digital village index, the number of e-commerce enterprises, and regional innovation and entrepreneurship index. The data used includes text data from the Ali Research Institute, converted into point data using the Amap open source platform, and socio-economic data from various sources.

The study reveals a three-stage development pattern for Taobao towns: (1) explosive growth (2014-2016), characterized by rapid expansion due to government support; (2) steady growth (2017-2018), focusing on quality improvement; and (3) lean development (2019-2021), characterized by refined industrial chains and model development. Spatially, Taobao towns exhibit a "T-shaped and three-center" pattern, decreasing from southeast to northwest, with rapid growth in central areas. Global Moran's I and General G index analysis confirms positive spatial autocorrelation and high-value clustering. LISA analysis reveals Low-High outliers, indicating spatial heterogeneity and development disparities. Standard deviation ellipse analysis shows a north-south distribution trend, with the center of gravity gradually shifting to the southwest. GeoDetector analysis identifies express business volume, GDP, and the regional innovation and entrepreneurship index as the most significant factors influencing Taobao town distribution, highlighting the importance of e-commerce infrastructure, economic development, and entrepreneurial environment. Other factors like cargo volume, digital village index, and the number of e-commerce enterprises also play significant, though less influential roles. Interaction detection shows that the development of Taobao towns is a result of combined effects of multiple factors.

The findings address the research questions by demonstrating the spatiotemporal evolution of Taobao towns in China. The "T-shaped and three-center" pattern reflects the interplay of national policies (coastal development, rise of central China, western development) and the characteristics of e-commerce. The shift in the center of gravity towards the inland suggests a potential reversal of traditional geographical constraints. The importance of express business volume highlights the crucial role of infrastructure in enabling rural e-commerce. The significant influence of the regional innovation and entrepreneurship index underlines the importance of fostering a supportive entrepreneurial environment for rural e-commerce to thrive. The findings provide valuable insights for policymakers, e-commerce companies, and rural entrepreneurs.

This study provides a comprehensive analysis of the spatiotemporal evolution of Taobao towns, revealing a three-stage development pattern and a "T-shaped and three-center" spatial distribution. Key factors influencing their distribution include express business volume, GDP, and the regional innovation and entrepreneurship index. The findings offer valuable insights for policymakers, businesses, and entrepreneurs seeking to promote rural e-commerce development. Future research could focus on more detailed micro-level analysis, exploring the impact of specific policies and interventions, and examining the sustainability and resilience of Taobao towns in the long term.

This study primarily relies on secondary data from the Ali Research Institute, potentially introducing biases. The analysis focuses on the provincial level, limiting the understanding of the micro-level dynamics within individual Taobao towns. Future research should incorporate more granular data, field studies, and qualitative methods to provide a more nuanced understanding of Taobao town development. The model used may not capture all the complexities of the interactions between factors.