Abstract: In the era of digital economy, the innovation process of firms not only involves the Research and Development of new technologies and products but also involves the allocation of R&D resources by the new technologies and products. The Chinese government attaches great importance to the development of big data. Under the dual impetus of technological revolution and policy support, can Chinese enterprises apply big data to decrease misallocation in firm innovation? Big data refers to the integrated information assets derived from the circulation data containing massive information after being analyzed and processed by digital technology. Less literature directly studies how big data affects misallocation in firm innovation. Drawing on the ideas of Hsieh and Klenow (2009)’s productivity misalignment model, this paper constructs a model of enterprise innovation resources misallocation, expounded the impact of big data application on innovation resources misallocation from the theoretical level, and verified the inferences of the theoretical model by using the data of China listed firms from 2011 to 2020. It is found that 52.83% of industries in China have an insufficient allocation of R&D capital and 62.26% of industries have an insufficient allocation of R&D staff. The results show that big data significantly restrains the misallocation of R&D capital and R&D staff, and this inhibition effect is more obvious when R&D resources are insufficient. The possible reason is that big data application contributes to the information collection, information processing, information integration, and information analysis between the internal R&D process and the external market environment of the enterprise, thus transferring to the allocation of R&D factors. For example, in the capital market, big data application can solve the problem of insufficient supply of traditional finance by expanding financing channels and accurately allocating credit resources, and breaking the inefficiency of resource allocation in traditional financial markets. In the labor market, big data application can eliminate the incompleteness of information to a certain extent to make the supply and demand information of employees in enterprise R&D services more accurate. The conclusion still holds after a series of robustness tests. The mechanism test shows that big data application affects the misallocation of firm R&D resources mainly through the knowledge flow effect and the technology barrier effect. On the one hand, big data application can produce a knowledge flow effect, accelerate knowledge absorption and knowledge diffusion between enterprises, and improve the mismatch of innovation resources. On the other hand, although big data application can break through the technical barrier effect of myopic investment by management, it is not conducive to improving the allocation of innovation resources for enterprises with low employee skills. Furthermore, the resource complementary effect indicates that the effectiveness of data elements also needs the support of technical conditions. When enterprises effectively promote the construction of digital infrastructure, the operational efficiency of enterprises can be improved, and the information analysis and decision-making optimization of the innovation process can be realized at a smaller cost, resulting in the efficient allocation of innovation resources. The conclusion of this paper provides useful enlightenment to promote the rational allocation of R&D resources and speed up the construction of an innovative country with data resources as the key element under the background of new scientific and technological revolution.