The second scenario uses data capture software that identifies common data elements across multiple documents and maps them to columns in your Excel, CSV or database output. What they can't do well is data validation or appending multiple documents to a single spreadsheet to build a dataset. The output would be one Excel or CSV file per document, and will usually require a bit of clean-up to remove extra text on the document that isn't part of the table. These applications can convert standard table data to individual spreadsheets. The first scenario can be managed using desktop OCR applications like ABBYY FineReader, ReadIRIS and Kofax OmniPage. Large batches of documents with complex tables or separate fields combined to make a table.
Converting scanned documents or PDF files to Excel, Google Sheets, CSV and other spreadsheet formats usually involves one of these two scenarios.
