During the last decade, the source to tap risk-based approach to pathogens in drinking water has been largely promoted. This paper addresses the issue of source water pathogen contamination, which is the first step of quantitative microbial risk assessment. It is focused on a selection of pathogens considered to be a major risk to human health. Source water quality is highly variable and understanding the reasons for this variability is important as it will influence the requirements for treatment, treatment efficiency and the resulting health risk associated with the finished water. A framework for source water microbial quality assessment based on catchment surveys and monitoring programmes was set and tested on ten water sources. The monitoring programmes included faecal indicators and pathogens, during both baseline and hazardous event conditions. Concentrations varied greatly within and between systems. Faecal indicators were shown to be poor surrogates for pathogen presence and concentrations. There was no recurring evidence that the pathogens correlated together and links between microbial parameters appeared to be very site specific. Such variability between systems shows the importance of running local monitoring programs for use in risk assessment. Finally, pathogen detection methods are not yet optimal due to their sensitivity and to the lack of knowledge on viability and infectivity of pathogens. A great effort needs to be made in the future to ensure better quality data as this may have large implications in the statistical risk assessment calculations.