The present system and method relate generally to the field of Robotic Process Automation, particularly to a form data extractor for document processing. The system and method relate to a form extractor for document processing using RPA workflows that can be easily configured for different document types. The form extractor includes a set of templates for identifying the document type (classification) and extracting data from the documents. The templates can be configured, i.e., by the user, by defining the fields to be extracted and the position of the field on the document. The form extractor is resilient to changes in the position of the template on a page, as well as to scan rotation, size, quality, skew angle variations and file formats, thus allowing RPA processes to extract data from documents that need ingestion, independent of how they are created.