As a consequence of the large number of masonry buildings and their important historical value, there is an ongoing effort to investigate their structural behaviour. This has led to a continuous development of new methodologies and techniques applied to masonry structures, including an increasing application of non-linear analyses. The considerable computational effort required by detailed models, e.g. Finite Elements (FE) models, highlighted the need to develop simplified modelling procedures. In this context, the well-known Equivalent Frame Method (EFM) has become very successful. This simplified approach is based on the assumption that the behaviour of the masonry walls is modelled with a system of mono-dimensional beam elements (pier and spandrel panels). The great simplicity of implementation and interpretation of the results contributed to the widespread application of the method in both the scientific field and the professional practise. Despite the huge effort devoted to improve the effectiveness of the EFM, its application can be considered really effective only for new or existing regular buildings. For these structures, in fact, the presence of regular configurations and good connection between walls and floors guarantee a “box-like” behaviour under horizontal (e.g. seismic) loads. The “box like” system is a necessary requisite to obtain the in-plane behaviour of the masonry walls well described by EFM. This method is more difficult to apply to existing irregular buildings and, in particular, in the case of old structures in historical centres. In these cases, irregular geometry, the presence of deformable diaphragms and the interaction with other structures in aggregate configurations are critical issues for the reliability of Equivalent-Frame (EF) procedures. A comparison between EF and FE models is deemed useful for investigating some of these uncertainties and for exploring the reliability of EF procedures in the case of existing masonry buildings. In [1], a parametric investigation was performed for several types of masonry regular buildings with different geometrical configurations and for walls affected by geometrical irregularities. The effects of the presence of irregularities on EFM accuracy are presented in this paper.