The importance of adventitious air leakage under normal operational conditions and its reduction in order to save energy is highlighted by the relvant building standards of many countries. This operational leakage is often inferred via the measurement of air permeability, a physical property of a building that indicates the resistance of its fabric to airflow. A building’s permeability is the measure of airflow rate through its envelope at a constant pressure differential of 50 Pascals. However, operational pressure differences are dynamic and typically an order of magnitude lower than 50 Pascals. Thus there is much uncertainty when using a value of permeability in an attempt to predict operational air leakage. Powerful simulation tools can model the ventilation rates found in a building in great detail, yet these complex modelling tools contrast with the much simpler tools that are used frequently to estimate annual energy consumption for space heating in dwellings. For example, some building codes assume a simple fixed relationship between air permeability measured at 50 Pascals and mean background infiltration during the heating season; the so-called rule-of-20. This paper evaluates afresh this rule-of-thumb. Firstly, a theoretical model of adventitious air leakage for a dwelling is presented. Secondly, the predictions of the model are compared against those of CONTAM, and AIDA, validated airflow analysis tools, for an identical building and environmental conditions. Thirdly, the model is used to predict the mean infiltration rate and the corresponding energy required to replace heat lost via air operation infiltration during the heating season for an apartment and a terraced house located in 14 different UK cities. Finally, the predictions of the model are used to develop a relationship between the adventitious air leakage under pressure, operational infiltration, and energy consumption during the heating season. The relationship is used to discuss the validity, accuracy, and applicability of the rule-of-20 and its use by simple modelling approaches such as the UK’s Standard Assessment Procedure.