Manufacturing companies advertise their manufacturing capabilities and services on their company website using unstructured natural language text. The unstructured capability data published on the web is a rich source of formal and informal manufacturing terms and knowledge patterns. Through systematic mining of a large collection of capability text, new semantic models and knowledge graphs can be extracted that can be used as the stepping stone of more formal ontologies. The objective of this research is to develop a framework for better understanding, analyzing, and summarizing manufacturing capability data that is available on the websites of manufacturing companies. The findings can support supply chain decisions and may result in the discovery of new trends and associativity patterns in the data. The focus of this paper is on demonstrating how visual analytics (VA) tools can be used for gaining insights into manufacturing capability and the associativity pattern among the capability entities labeled by various terms. A visual analytics system, named Jigsaw, is used for exploring the connections between various entities such as process, material, industry, and equipment across the documents in the experimental dataset.