To perform an international trial to derive alert and action levels for the use of quantitative PCR (qPCR) in the monitoring of Legionella to determine the effectiveness of control measures against legionellae. Laboratories (7) participated from six countries. Legionellae were determined by culture and qPCR methods with comparable detection limits. Systems were monitored over ≥10 weeks. For cooling towers (232 samples), there was a significant difference between the log mean difference between qPCR (GU l −1) and culture (CFU l −1) for Legionella pneumophila (0·71) and for Legionella spp. (2·03). In hot and cold water (506 samples), the differences were less, 0·62 for Leg. pneumophila and 1·05 for Legionella spp. Results for individual systems depended on the nature of the system and its treatment. In cooling towers, Legionella spp. GU l −1 always exceeded CFU l −1, and usually Legionella spp. were detected by qPCR when absent by culture. The pattern of results by qPCR for Leg. pneumophila followed the culture trend. In hot and cold water, culture and qPCR gave similar results, particularly for Leg. pneumophila. There were some marked exceptions with temperatures ≥50°C, or in the presence of supplementary biocides. Action and alert levels for qPCR were derived that gave results comparable to the application of the European Guidelines based on culture. Algorithms are proposed for the use of qPCR for routine monitoring. Action and alert levels for qPCR can be adjusted to ensure public health is protected with the benefit that remedial actions can be validated earlier with only a small increase in the frequency of action being required. This study confirms it is possible to derive guidelines on the use of qPCR for monitoring the control of legionellae with consequent improvement to response and public health protection.