The invention relates to a computer-implemented system and method for automatically utilizing data from a document. The method may comprise the steps of: receiving a document; automatically scanning the document to identify at least one data field in the document; proposing a document definition to a user based on the scan; receiving an acceptance or modification of the document definition from the user through the user interface; automatically extracting at least one data element and at least one data field from the document using the document definition; automatically searching an existing database for at least one data element or data field that matches the data element or data field extracted from the document; and storing a link between at least one of the extracted data element or data field and a data element or data field in the existing database.