Unstructured data research in business: Toward a structured approach
针对管理者在具体场景下如何选择非结构化数据源和处理方法的困惑，本文基于组织学习理论，提出一个包含问题识别、方案开发和问题解决的三步结构化框架，帮助管理者有效利用非结构化数据辅助决策。
Despite the unprecedented growth in both the volume of unstructured data (UD) and the associated methodological sophistication, there is a growing managerial need for a structured view of how to select data sources and methods given a specific use case or scenario. Handling UD is typically resource intensive, requires many steps, and involves high uncertainty, but UD can contain rich information not found in structured data. Recognizing the gap in clear guidelines for leveraging UD in managerial decision-making, we develop a systematic three-step approach: (1) problem identification, (2) solutions development, and (3) problem resolution. Building on organizational learning theory, we propose a solutions development framework with four conceptually distinct uses of UD based on two dimensions: organizational learning goals (exploration and exploitation) and environmental scanning scope (internal and external data sources). Finally, we discuss implications for practitioners and outline key focus areas for future research directions.