In order to explore the impact of Crowd-to-Crowd interactions on crowdsourcing innovation contribution, from the perspective of social network analysis(SNA) and based on the data derived from typical crowdsourcing innovation platform—Dell"s IdeaStorm, this paper analyzes the interactive network structure characteristics of Crowd-to-Crowd interactions under the circumstance of crowdsourcing innovation. Then an empirical study is conducted through taking the network central indicators (Out-Degree centrality, In-Degree centrality, Betweenness centrality, In-Closeness centrality and Out-Closeness centrality) to measure the Crowd-to-Crowd interactions, and taking the number of ideas submitted, scores, number of votes obtained to measure the crowdsourcing innovation contribution. We find that Crowd-to-Crowd interactions has a significant impact on the innovation contribution, out-degree centrality is the most positive factor for crowdsourcing innovation contribution. But not all the key variables have the significant positive impacts; in-degree centrality has negative influence on the number of ideas submitted, betweenness centrality has negative influence on the number of ideas submitted and scores. Meanwhile, closeness centrality negatively affects the crowdsourcing innovation contribution. It is necessary for designing a scientific interactive feedback mechanism and an effective reward scoring system to guide crowds to communicate more effectively for contributing high-level innovative solutions.