Achieving a transition to a low-carbon energy system is now widely recognised as a key challenge facing humanity. To date, the vast majority of research addressing this challenge has been conducted within the disciplines of science, engineering and economics utilising quantitative and modelling techniques. However, there is growing awareness that meeting energy challenges requires fundamentally socio-technical solutions and that the social sciences have an important role to play. This is an interdisciplinary challenge but, to date, there remain very few explorations of, or reflections on, interdisciplinary energy research in practice. This paper seeks to change that by reporting on an interdisciplinary experiment to build new models of energy demand on the basis of cutting-edge social science understandings. The process encouraged the social scientists to communicate their ideas more simply, whilst allowing engineers to think critically about the embedded assumptions in their models in relation to society and social change. To do this, the paper uses a particular set of theoretical approaches to energy use behaviour known collectively as social practice theory (SPT) - and explores the potential of more quantitative forms of network analysis to provide a formal framework by means of which to diagram and visualize practices. The aim of this is to gain insight into the relationships between the elements of a practice, so increasing the ultimate understanding of how practices operate. Graphs of practice networks are populated based on new empirical data drawn from a survey of different types (or variants) of laundry practice. The resulting practice networks are analysed to reveal characteristics of elements and variants of practice, such as which elements could be considered core to the practice, or how elements between variants overlap, or can be shared. This promises insights into energy intensity, flexibility and the rootedness of practices (i.e. how entrenched/ established they are) and so opens up new questions and possibilities for intervention. The novelty of this approach is that it allows practice data to be represented graphically using a quantitative format without being overly reductive. Its usefulness is that it is readily applied to large datasets, provides the capacity to interpret social practices in new ways, and serves to open up potential links with energy modeling. More broadly, a significant dimension of novelty has been the interdisciplinary approach, radically different to that normally seen in energy research. This paper is relevant to a broad audience of social scientists and engineers interested in integrating social practices with energy engineering.