Understanding Benefits and Limitations of Unstructured Data Collection for Repurposing Organizational Data

Abstract

With the growth of machine learning and other computationally intensive techniques for analyzing data, new opportunities emerge to repurpose organizational information sources. In this study, we explore the effectiveness of unstructured data entry formats in repurposing organizational data in solving new tasks and drawing novel business insights. Unstructured data accounts for more than 80% of the organizational data. Our research analyzes the implications of using unstructured data entry formats for propagation of organizational styles. We study this phenomenon in the context of case management in foster care. Using natural language processing and machine learning, we show that unstructured data formats foster entrenchment and propagation of individual organizational styles and deviations from the industry norms. Our findings have important implications both to theory and practice of business analytics, conceptual modeling, organizational theory and general data management.