Scanned files often contain multiple documents within. For example, a mortgage loan file package contains up to 100 documents. An insurance claims file may contain forms, invoices, receipts, ID documents. A medical file could contain dozens of documents from patient intake forms to X-rays. A box of records might contain hundreds of diverse documents.
For a digital process to make sense of the information contained within that file, each document within it must also be separated. This is a precondition of document classification and data extraction.