Impact Management Assessment Services parse imported documents to extract and store metadata. Metadata includes sections that are in the document; tables and columns that are used by each data model, query, and results section; and section dependencies (for example, Results A depends on Query B depends on Data Model C).