This article introduces the Hyperbole Identification Procedure (HIP), a first systematic method for identifying linguistic hyperbole in discourse. We start by comparing existing definitions of linguistic hyperbole. Based on the commonalities shared by these definitions, we provide our operational definition of hyperbole as “an expression that is more extreme than justified given its ontological referent.” The next section argues why it is useful to identify hyperbole, as with metaphor in Metaphor Identification Procedure Vrije Universiteit (MIPVU), at the level of lexical units, and subsequently introduces the steps of HIP. We follow up with two sample analyses of HIP in practice. First, we show how to unitize and analyze one complete sample sentence. Second, we present sample analyses of a number of selected cases. Then we present data showing that HIP can be reliably applied to a sample corpus of Dutch news texts. We end with discussing applications and implications of using HIP in corpus research.