Graphs effectively represent structural relationships, while time series capture temporal dynamics, both of which are critical to understanding complex systems, such as asset management, IoT optimization, and micromobility demand predictions. In these contexts, the interplay between evolving entities and their relationships, captured by graphs and large volumes of time-series data, remains challenging to fully exploit, due to the absence of a unified approach. Practitioners are thus forced to treat both data structures as isolated and must create connections with manual effort. Our vision, HyGraph, includes a hybrid data model and operator concept designed to integrate the expressive power of temporal graphs with time-series analysis, providing a holistic approach for complex queries, analytics, and predictive tasks, which are currently unfeasible by working solely on isolated data structures. This vision has the potential to drive significant advancements in both research and practice, addressing limitations associated with isolated data models and fostering new opportunities for interdisciplinary insights.