Abstract The building sector contributes a major proportion of the global energy consumptions and carbon emissions. The energy performance or efficiency of buildings can be improved through a wide range of retrofitting measures which can have very different costs. Under budget, time and other resource constraints, it is not practical to apply all energy saving measures to a given retrofitting project. As such, there is a need to rank and select the most cost-effective measures to meet efficiency improvement goals. Traditionally, energy efficiency improvement measures and their costs are evaluated separately which makes prioritising among the measures difficult. In response, an integrated approach by soft-linking a metamodel-based Bayesian method and a life cycle cost assessment method is proposed to rank and select the most cost-effective retrofitting measures. The metamodel-based method is used to compute building energy consumptions before and after retrofit; and the cost-assessment method is used to evaluate the life cycle cost of implementing each measure. A selection of nine retrofitting measures are ranked according to life cycle energy savings, life cycle cost, and cost-effectiveness (measured by cost per unit energy saved). Findings from the Singapore case study suggest that retrofitting building envelop is the third least cost-effective measure although it can lead to highest energy savings. Lighting replacement has the least life cycle energy savings, but it is the most cost-effective measure. Electricity price has little influence on the cost-effectiveness ranking of all nine measures but discount rates (tested for 4%, 7% and 12%) can influence the ranking of home appliances. Based on the findings from the case study, the proposed integrated approach can help identify an optimum retrofit strategy and the cost of achieving energy efficiency targets for existing buildings.