Insomnia Disorder is the most prevalent sleep disorder, and it involves both sleep difficulties and daytime complaints. The neural underpinnings of Insomnia Disorder are poorly understood. Several existing neuroimaging studies focused on local measures and specific regions of interests, which makes it difficult to judge their whole-brain significance. We therefore here applied a data-driven approach to assess differences in whole-brain structural connectivity between adults with Insomnia Disorder and matched controls without sleep complaints. We used diffusion tensor imaging and probabilistic tractography to assess whole-brain structural connectivity, and examined group differences using network-based statistics. The results revealed a significant difference in the structural connectivity of the two groups (p = .014). Participants with Insomnia Disorder showed reduced connectivity in a sub-network that included mainly fronto-subcortical connections with the insula as a key region. By taking a whole-brain network perspective, our study enables the integration of previous inconsistent findings. Our results reveal that reduced structural connectivity of the left insula and the connections between frontal and subcortical regions are central neurobiological features of Insomnia Disorder. The importance of these areas for interoception, emotional processing, stress responses and the generation of slow-wave sleep may help guide the development of neurobiology-based models of the prevalent condition of Insomnia Disorder.