Currently, the term innovation is often associated with the generation of new technologies, excluding other forms of innovation, with equal or greater value, such as social innovation. Technological innovation has been repeatedly used as a means differentiation in order to achieve a higher enterprise competitiveness. This has led to achievement of great scientific and technical progress. Nevertheless, in the last few decades, this has not prevented some territories from social issues such as poverty, exclusion, inequality, deterioration of environmental conditions and unemployment, among others. All these problems have a high local impact, being here where social innovation importance as a force for potential change is growing. Despite the broad consensus regarding the potential and the positive effects that social innovation has in those territories, there is still considerable debate surrounding how this concept should be defined and what factors contribute to its appearance. There are many variables that have been mentioned in the literature; however, a review of previous works shows that it is an open field for the identification of new ones. As described above, this paper has three main aims: to analyze the state of the art up to now, to give a definition of social innovation and to identify, analyze and weigh those key factors for its emergence in the territories. To this end we have designed a methodology based on a series of qualitative analysis techniques. Firstly, a review of the state of the art has been carried out to obtain a first approximation to the concept of social innovation and determine which elements are decisive in its appearance. Secondly, based on the existing literature and knowing what the problem is, we complete the methodology by designing and conducting semi-structured interviews. We chose this technique because we understand that social innovation is a living and constantly evolving phenomenon. This makes it very useful and allows us to confront the ideas and evidence from our literature review in a pragmatic way. The information gathered from the interviews allow us to identify, in a reliable and empirical way, new key elements to define social innovation. Additionally, the interviews help us to recognize and weigh the existence of new essential factors for the development of social innovation. These interviews have been conducted with a total of 24 agents involved in social innovation processes: managers and directors of centers responsible for promoting social innovation; agents that developed socially innovative initiatives; and scientists and academics who published work in this field. To analyze the content of the 24 interviews, we used the Atlas.ti computer program. This tool helped us analyze the transcripts. In order to explore and interpret the texts we analyzed 24 primary archives, which represent one transcript for each interview carried out. Afterwards, we encoded them. This process is a form of data reduction, since, based on a large volume of information, we managed to segment certain pieces of information (quotes) and label them based on codes. These are the main elements of analysis and they are considered as a system of categories. In our case, each time we have identified a quote, which made reference to the existence of a factor that conditioned the irruption of social innovation or that enunciated characteristics that defined it, we labeled a code-referred to a certain factor related to the content of the quotes-. Once the individual coding and data editing was completed, we organized the codes into families. Specifically, in line with the main purposes of our analysis, in two families: one to collect the codes related to the definition of social innovation and a second to gather those related to its determinant factors. To provide our study with greater depth, it was essential to carry out a statistical analysis of the content that would allow us to know the frequency of appearance of the codes. First, we calculated the absolute frequency of the codes, referred to the number of times a quote was identified with the same code. Nevertheless, to obtain more information about the importance of each one of the codes, we relativized them considering that they had previously been divided into families. Thus, we have obtained the relative frequency by families, which is, the number of times any code is repeated within a family according to the total number of times codes included in that family appear. This contributes directly to know which factors have more strength in the emergence of social innovation, among the ones pointed out by the interviewers. After the performance of the aforementioned analysis, we suggest the following definition of social innovation: Practices or initiatives carried out by the community that, based on the products, services or models they generate, or through the process designed to achieve their objectives, they reach solutions that give a better response, in an alternative and creative way, to social problems or needs. In addition, we point out five key points for the development of social innovation: social and cultural elements; political and institutional support; knowledge and facilitating mechanisms; spatial components and entities; and mechanisms that determine the corporate and social productive structure. These gather 29 factors identified and weighted according to the number of times they were indicated by the interviewees. Therefore, this work, not only points out the key factors of social innovation and the foundations on which they are grouped by theme, but also shows the importance of each underlying determinant according to its absolute and relative frequencies. These frequencies are used as a proxy to weigh the importance of these factors and foundations. This allows to know which ones are more needed and have more weight in the emergence of social innovations. The ten factors with more impact are: the socioeconomic level of the population (0.09532), the existence of strategies and lines of action to develop social innovation (0.08838), the existence of laboratories or centers of social innovation (0.05892), the existence of public funding (0.05892), the existence of social concern (0.05719), the cultural propensity to change (0.05719), knowledge of the local reality (0.05545), knowledge of the opportunities and potential of new means and instruments (0.04506), links encouraging public-private relations (0.04332) and the existence of cooperatives (0.03986). In summary, it should be noted that the recognition of these factors, together with the weighing of their relative importance, represents a considerable innovation with respect to the existing texts. This work has allowed us to verify the suitability of some determining factors that had previously been pointed out in other studies, state new ones that emerged after the investigation, explain the correlation between them and present how important and what role each of them plays to have a beneficial ecosystem for social innovation. This knowledge can be very useful when establishing lines and criteria for the design of prioritization strategies, useful where there are restrictions of a political, organizational or budgetary nature.
