Due to the ongoing adoption of new household energy technologies, there has been an increasing effort to analyze these behavioral patterns and predict future dynamics in space and time. Likewise, power system planners and policy makers aim to assess the effects of large-scale adoption of energy technologies. This paper presents the principal data sources and building blocks for spatiotemporal models that can be used to simulate the large-scale adoption of distributed energy resources. We explain the advantages of the estimates/forecasts obtained by these methods and limitations modelers and decision-makers face during implementation. Finally, we provide an overview over the various use cases for such models in power systems, outlining the advantages and barriers of each model typology for these domains. This comparison shows a trend towards computationally heavier and more complex models-based approaches rooted in Artificial Intelligence, such as Agent-based Modelling, mainly applied to consumer characterization or electricity distribution network planning.