Human capital and a transformed data set are combined into a stochastic frontier analysis to estimate the technical efficiency of star-rated hotels in 30 provinces and municipalities of China for the period of 2011-2015. The influences of revenue diversification across room, food & beverage and other services and incoming tourists on technical efficiency, labour and capital productivities are addressed. Hotels in Zhejiang, Shanghai and Tianjin are found to be the most efficient. For the efficiency and productivities determinants, per capita GDP of the hosting province is the primary determinant, hotel ranking and diversification in revenue sources contribute to efficiency and partial productivities, while tourists from different origins bring about diversified impacts. Hotels located in regions with more Taiwan tourists are more efficient with higher capital productivity. Tourists from Hong Kong, Macau and foreign countries have brought about both positive and negative linkages to efficiency and partial productivities. In general, hotels can reallocate their resources to diversify their services for better efficiency and productivities. In contrast, expanding external markets cannot guarantee higher efficiency and better productivities. Hotel management should be careful in formulating their marketing strategies.