Abstract: A survey analysis process is disclosed for extracting and organizing dynamic textual content to use as input to structural equation modeling (SEM) for survey analysis in order to understand how customer experiences driver customer decisions. The survey analysis process applies natural language processing (NLP) to understand customers' open-ended comments and segment customers based on the topics. By doing this, open-ended questions can be handled in SEM. The survey analysis process then builds a structural equation model to evaluate how customers' real feelings affect their loyalty, satisfaction score, etc., which are measured as closed-ended questions in the survey. In this way, the survey analysis process enables SEM to handle customers' actual feelings.