The development and success of the ‘business analytics’ in the private sector, in combination with the growing availability of large quantities of useful data in government agencies, gives rise to the emergence of the ‘policy analytics’ in the public sector. However, though some knowledge has already been developed in this area, extensive research is required in order to increase our knowledge base concerning the exploitation of these exponentially increasing quantities of data available in government, in combination with data from private sector firms as well, using advanced analytical techniques (from various areas, such as machine learning, statistics, simulation, etc.), in order to provide substantial support for all stages of public policies in various important policy domains. This paper makes a contribution in this direction, by describing a methodology for policy analytics in the economic policy domain, concerning a highly important problem: the economic crises, which repeatedly occur in market-based economies being an inevitable trait of them. Our methodology aims at the identification of firm’s characteristics that affect positively or negatively their sensitivity to the economic crisis, which enables a deeper understanding of the kinds of firms that exhibit higher sensitivity to economic crisis (i.e. have more negative consequences) and provides a basis for the design of public policies for supporting such firms. It exploits existing data from various public sources (e.g. Ministries of Finance, Statistical Authorities), in combination with data from private sources (e.g. business information firms, consulting firms), from which firm-level crisis sensitivity models are estimated. Furthermore, an application of the proposed methodology is presented, using data from Greek firms for the crisis period 2009 – 2014, which provides interesting insights.