The most widespread application of learning analytics is the early identification of students predicted to be at risk of failure or withdrawal. Predictive models can be built using historical data about the activity of previous students. These suggest, often with a high degree of accuracy, whether a current student is at risk or not, based on that student’s own patterns of activity. This ‘actionable intelligence’ can then lead to an intervention of some kind with students in an attempt to change their behaviour and improve their chances of academic success. The financial implications of reducing attrition make investment in early alert systems extremely attractive to senior management in institutions where dropout is a serious problem.