Abstract:     With the development of digital economy, the innovation activities of enterprises have covered the external network, and the improvement of innovation performance is always the goal pursued by enterprises. The mismatching of network dual embeddedness with digital innovation search strategy is serious. There are such frontier problems as how to carry out two-dimensional matching between network dual embeddedness and digital innovation search strategy, and how the matching influences performance of open innovation in the field of innovation. The matching that is helpful to optimize and allocate resource becomes one of the important sources of open innovation performance. Under the condition of limited resources, it is urgent to solve the mismatching and improve performance of innovation.
   This paper based on the perspective of two-dimensional matching, has explored two-dimensional matching between network dual embeddedness and digital innovation search strategy, and analyzed the impact of matching on open innovation performance. It constructed patterns of network dual embeddedness and digital innovation search strategies. Data were collected via questionnaire survey, and valid questionnaires were obtained from 209 enterprises. The data was analyzed with the methods of SPSS and AMOS, and it was also tested empirically with the methods of ANOVA, independent-samples T test and other methods. 
    Results of the research show matching of network dual embeddedness (strongly relational, highly structural) with digital innovation search strategies (highly exploitative, highly exploratory), network dual embeddedness (weakly relational, highly structural) with digital innovation search strategy (lowly exploitative, highly exploratory), network dual embeddedness (strongly relational, lowly structural) with digital innovation search strategy (high exploitative, lowly exploratory), network dual embeddedness (weakly relational, lowly structural) with digital innovation search strategy (lowly exploitative, lowly exploratory). Moreover, matching of network dual embeddedness with digital innovation search strategy positively affects open innovation performance.
It is of great significance to explore the matching of network dual embeddedness with digital innovation search strategies. Theoretically, this paper extends the study of network dual embeddedness and digital innovation search strategies to the level of two-dimensional matching. It makes up for the shortage of literatures, overcomes the limitations of matching, improves the theory, enriches the research of matching. Practically, this paper provides a basis for strategic decision-makers to choose the appropriate digital innovation search strategy, and a new direction for enterprises to improve performance of open innovation in the same cost. Enterprises can take two-dimensional matching as an effective way to improve the performance of open innovation. Managers may adopt appropriate combination of digital innovation search strategies according to the degree of network dual embeddedness.
    There are still some limitations in the study. Firstly, when exploring the influence of two-dimensional matching patterns on open innovation performance, comparison of the four matching patterns is not made. Comparative studies can be carried out in the future; Secondly, although four two-dimensional matching patterns are constructed and tested, it is not explored which pattern should be adopted by enterprises in different development stages. In the future, case studies can be conducted to explore the correlation mechanism between two-dimensional match patterns and the development stage of enterprises, and reveal the matching law of enterprises in different stages.

Key words:  network dual embeddedness, digital innovation, two-dimensional matching, open innovation performance