The growth of available data in the social sciences led to numerous knowledge discovery projects being launched over the years. Even if the volume and the speed of data are increasing, in social sciences data has an important limitation in terms of methodological process that drives the conceptual and analytical questions posed to the data. Social sciences domain experiences several challenges in their desire of extracting useful and implicit knowledge due to its inherent complexity and unique characteristics, as well as the lack of standards for data mining projects. The aim of this research is to bring Cross-Industry Standard Process for Data Mining (CRISP-DM) methodology as a standardization in analyzing large volumes of unstructured data to generate analytical insights for wellbeing and social sciences topics in general. Also, taking into consideration that for a data scientist, the most time-consuming activity is data preparation step, we are trying to make more efficient this process using a clear methodology and tasks. Conclusion is that using a strong methodology with well-defined steps in research can increase productivity in terms of time and enhance the quality of the research.