Abstract Research on the progressive aspect in Germanic and Romance languages has benefited from corpus data. A transparent, objective, reliable and replicable identification of such constructions in corpora is however challenging. The present chapter presents preliminary methodological work in automatically retrieving and counting authentic examples from treebanks, that is, grammatically annotated corpora. It demonstrates how selected constructions that mark the progressive in Italian and Norwegian are collected from treebanks accessible through the INESS platform. Deep syntactic relations such as those between predicates and arguments are factored in and quantified. Corpus queries that exploit syntactic dominance relations are potentially more powerful than queries using only linear precedence, but there is a relative shortage of treebank resources.