This research project investigates the influence of a number of economic and financial variables on the profitability of Greek enterprises from 2006 to 2013, namely: annual revenue, exports as a binary variable, number of employees, sectoral composition, investments and founding year. Three models were tested using linear stepwise regression as well as logistic regression. The fit to the data is low, indicating that other important factors, in addition to those tested, influence the development of corporate profits more. Running the models for individual years we see that in the years 2010 and 2011 mark a shift in the direction of the relation of several of our variables to profits (such as the age of the enterprise and the number of employees). The full application of austerity policies because of the crisis may be the explanation for that, indicating that the crisis did not have a unified influence on profitability. More importantly statistically significant results are observed in relations which conflict with economic theory, namely exports are negatively correlated to profitability (Model 1) or not correlated at all (Model 2). Our models do not contribute to understanding the determinants of profitability but rather the difficulties to identify smooth profitability patterns at times of crisis and austerity. The research is still in progress as we strive to improve the explanatory power of the models, by expanding the number of variables and using additional statistical methods.

The present paper tries to devise a rating system for evaluating the likelihood of success of publicly supported companies. In recent literature we encounter quality indexes, composed of variables like name selection, registration location, intellectual property rights etc., which predict fairly accurately startup success. Unfortunately such indexes can only be compiled for countries with open administrations facilitating access and cross-checking of company data. We argue, however, that it is countries with less open administrations that have the highest need of ex ante quality assessment in order to maximise the public return of investment of state support mechanisms. This is why we suggest that, rather than giving up research on the likelihood of company growth because of lack of systematic open data, we can test firm dynamism based on public domain information. Dynamic firms, we assume, will be more likely to grow. The Web and Social Media data, which have until now mostly been used for marketing purposes, can be a valuable source of information for our purpose. We attempted to use the digital footprint as a proxy for the dynamism of SMEs and startups. Using State Aid schemes (i.e. public support to profit-oriented companies) in the period 2007-2013 in Greece we constructed a sample of 2000 companies and rated them individually for their presence (or not) in Facebook, LinkedIn, a site (with or without registered domain name), e-mail with domain name, references, size of the site, sales via web, search function, communication, regular update, FAQs and after sales support availability. Testing for sector, region, type of company and type of support scheme we find a statistically significant result for the more dynamic sectors, type of schemes and type of companies, while the regional level appears to be irrelevant, as companies in all regions demonstrate a fairly unified digital presence.