Descriptive statistics of relevant variables.

The development of Happy Farmhouse tourism in China faces significant challenges of homogenized competition and low operational efficiency. This study aims to evaluate the operational efficiency of these farmhouses and compare the performance across three distinct business entity models: individual farmer, farmer participation, and company cluster. To this end, the study employs a three-stage Data Envelopment Analysis (DEA) model on panel data (2014–2023) from 50 farmhouses in Huzhou, China, to measure technical, scale, and managerial efficiency. Results reveal steady improvements in overall performance but significant efficiency disparities, with company clusters demonstrating the highest efficiency, followed by farmer participation models, and individual farmers demonstrating the lowest. The primary reasons for this disparity are identified as differences in scale efficiency and talent supply. Consequently, our findings provide actionable insights for optimizing rural tourism’s role in sustainable development and rural revitalization, with policy recommendations emphasizing talent recruitment, cluster-based management, and targeted support for different business models.