As far as the dataset is concerned, a novel but tailored dataset will be constructed by drawing from various sources in order to map Greek start-ups. In particular, multiple datasets will be used, like the dataset provided from the population of Greek start-ups listed on Elevate Greece, an initiative launched by the Greek government to identify promising start-ups and therefore a leading provider of information regarding start-ups’ access to external financing in Greece. However, as with any other dataset, Elevate Greece does not capture the same amount or type of information for all start-ups. Therefore, additional datasets will be used, such as Crunchbase (an online directory for start-ups), data provided by digital hubs, such as COSMOTE InnovatiON and Orange Grove, and information given by the General Commercial Registry (G.E.MI.). Additionally, secondary sources will be used (e.g., corporate websites and social platforms) to confirm the analysis regarding the information about the founder’s characteristics. Moreover, fine-grained information for academic founders will be collected from Scopus, university websites, and other sources. To further enhance the dataset, firm-specific information will be obtained, namely, financial data and information about firms’ profiles, from the Bureau van Dijk’s Orbis database and each firm will be visually inspected one-by-one to document each firm’s capital structure and discern funded firms from unfunded ones.

As far as the methodology is concerned, the aim is to establish outranking relations among the start-ups by evaluating them on a set of conflicting criteria (which correspond to quantitative and qualitative firm specific characteristics) that affect their access to external capital. Based on multi-criteria approaches, the proposed methodology will distinguish between financed and non-financed ones. The performance of such a non-parametric approach can overcome the shortcomings of the statistical and econometric techniques, such as the difficulty in explaining their parameters, and the difficulties often encountered in the parameters’ estimation procedure. The development of the proposed model will also be based on collaboration with experienced expert investors, which will allow capturing the decision-making policies of actual investors.