This study involves a detailed analysis of a mechanically ventilated heat recovery (MVHR) system installed as a case study retrofit project in a UK test dwelling. Comparisons of predicted and in situ performance are made through the calculation of a theoretical and practical heat loss coefficient value (HLC). Analysis revealed underperformance in the installed MVHR system, as compared to predicted data. Issues were identified in the installation and function of the control unit and ductwork. Improvement work enabled the system to function more closely to optimum performance levels. However, discrepancies still existed between the predicted and actual HIE values, which were largely accounted for through theoretical heat loss calculations attributable to slight imbalances within the system supply and extract rates. When normalised, the data still showed an unaccounted heat loss of approximately 6 W/K in both MVHR tests. The study reinforces the need for high quality MVHR installation and commissioning work, which is critical to ensure that a system can perform to the optimum levels specified in manufacturer literature. Education and training throughout the supply chain is essential in achieving this, as system inefficiencies can lead to unnecessary heat losses resulting from increased levels of air infiltration or leakage. (C) 2015 Elsevier B.V. All rights reserved.