High-profile terrorist attacks in the new millennium have made counter-terrorism a priority for many governments. The abundance of open source intelligence available after the 9/11/2001 attack, for example, enabled a comprehensive analysis of that terrorist network. The opportunity, thus, is to collect valuable and cost-effective information during the pre-attack phase to aid in the prevention of such atrocities in the future. This article positions social network analysis as a key tool for this type of intelligence analysis, with emphasis on the automated extraction of data relevant to the structural organisation of its actors and the attributes of their relationships in the network by using data and text mining techniques on open sources. The processes outlined are viewed as a layer that could complement and help to populate a terrorist behavioural activity model such as where the recognition of pre-incident indicators are linked to the likelihood of terrorist events.