“Although activation services such as monitoring, training, job subsidies or workfare have been shown to increase exits from unemployment, there is no evidence about their effects during recessions. We address this policy-relevant question by evaluating a large activation programme introduced in Portugal in early 2012, a time of very high and still increasing unemployment. The programme was based on requiring specific unemployment benefit recipients to meet caseworkers in jobcentres and then participate in active labour market policies. Our analysis draws on rich longitudinal data, the targeted nature of the programme (namely of its component focused on those unemployed for at least six months), and fuzzy regression discontinuity methods. We find that, despite the weak labour market, the programme is very succesful as it doubles the monthly reemployment probability. Moreover, we find no effects in terms of income or transitions to non-employment. The results are robust to a number of checks, including a falsification exercise based on pre-programme data.“