Chinese E-commerce market is increasingly saturated, fierce competition has spawned a new E-commerce model: more and more enterprises set up direct-sale stores on online shopping platforms to reduce costs by lessening intermediate links of sales and make products more competitive. However, new model also makes the enterprise face a series of customer needs from production to after-sales. Quality function deployment (QFD) has been proved be an effective method to integrate customer requirements (CRs) into the improvement of engineering characteristics (ECs) of products or services. To effectively identify CRs and determine the development direction of ECs, this article proposes a new QFD method based on the extended Kano model and social network analysis (SNA) method by considering the characteristics of online shopping reviews and the social relationship between the enterprise's departments. CRs are identified through crawler and word frequency analysis, while their Kano categories are generated by sentiment analysis and frequency-inverse document frequency algorithm. The objective weights of CRs are determined by word frequency and dynamically adjusted depending on enterprise's development stages. A SNA based consensus reaching process is devised to determine ECs’ priorities. The practicality of the proposed QFD model is verified by a case study of mobile phone design under the new E-commerce model. A comparative analysis shows the essential characteristics of the model, and the influence of calculation parameters is shown by the sensitivity analysis.