In Brazil, official data confirms that residential vacancy is bigger than the housing deficit. While the simultaneous existence of a built empty stock and families in need of housing has been widely criticised, very little is known about the vacant stock itself. It is precisely the lack of information one of the main obstacles to face this controversial scenario. This dissertation contributes to the scientific discussion on real estate vacancy by presenting a method to visualise the phenomenon in Belo Horizonte, Brazil, using geographic electricity consumption data, which discloses the location of utility poles, characterised by the number of connections in each pole and their billing status. First of all, the literature review discusses how the capitalist production of space generates a vacant built stock and, afterwards, how the scientific community have worked with unconventional data sources to inform real estate vacancy. Next, the gathered data is analysed and, from that, Geographic Information Systems (GIS) support the construction of a Geoprocessing model. The process results on the visualisation of areas of interest where there is a high absolute density of possibly vacant properties. The method is firstly applied at the municipal scale, and the broader scenario of the city, as well as the method’s limitations, are discussed. Then, the process is applied at a local scale, focusing on the Centre neighbourhood. After that, it is conducted a preliminary test of its effectiveness through the direct sampling method; photos of the identified areas of interest are collected and analysed. Finally, conclusions are presented, also revealing how the developed method could serve as a tool for public authorities to monitor real estate vacancy.