The ability to estimate the probability of default has always been one of the main goals of financial institutions and, in the near future, will certainly be the key to coping with the increase in non-performing exposures due to the impacts of the Covid-19 pandemic emergency. This paper aims to investigate the credit risk management practices and tools adopted to detect the company’s economic and financial distress at an early stage. Accordingly, an in-depth review of the insolvency prediction models was first conducted to illustrate limitations and strengths widely discussed in the literature. The case study analysis was subsequently used to describe both the Early Warning System (EWS) methodology and the internal governance practices adopted to monitor and promptly manage signs of deterioration in creditworthiness. The results highlights the necessity to combine quantitative models with the judgmental approach in order to set up an effective monitoring process capable of intercepting anomalies and avoiding misclassification. The case study analysis also shows how the early involvement of debt restructuring experts is a key element in defining actions to prevent default events.