In this paper, we present a deployed prototype of a scalable low-cost solution providing personalised home heating advice to households. Our solution, named MyJoulo (www.myjoulo.com), uses intelligent algorithms to analyse data collected from a specially designed USB temperature logger, placed on top of the thermostat, in order to build a thermal model of the home and to infer the operational settings of the heating system. This model is then used to calculate the impact, in terms of percentage reduction in heating costs, of various interventions (such as reducing the thermostat set-point temperature or adjusting timer settings); providing specific actionable advice to the household. The system was launched in beta form in December 2012 and registered over 750 users in its three months of operation.