While the literature on theoretical macroeconomic models adopting the stock–flow consistent (SFC) approach is flourishing, few contributions cover the methodology for building an SFC empirical model for a whole country. Most contributions simply try to feed national accounting data into a theoretical model inspired by Godley/Lavoie (2007), albeit with different degrees of complexity. In this paper we argue instead that the structure of an empirical SFC model should start from a careful analysis of the specificities of a country's sectoral balance sheets and flow-of-funds data, compared to the relevant research question to be addressed. We illustrate our arguments with examples for Greece and Italy. We also provide some suggestions on how to consistently use the financial and non-financial accounts of institutional sectors, showing the link between SFC accounting structures and national accounting rules.