How the Table Reconstruction Works
A PDF stores text as fragments positioned by exact coordinates, not as a grid of cells β which is why copying from a PDF so often scrambles a table. This tool reads the position of every text fragment, groups fragments that sit on the same line into rows, and orders them left to right into columns, rebuilding a spreadsheet that mirrors the page. Each page of the PDF becomes its own worksheet so nothing is jumbled together.
It works best on 'digital' PDFs that carry a genuine text layer β exports from accounting software, bank statements, reports and the like. The reconstruction is automatic and best-effort: very complex or irregular layouts may need a little cleanup in Excel afterwards. A scanned PDF (a photo of a page) has no text layer, so run it through the OCR tool first, then convert.
Processed in Your Browser
The text is extracted and the workbook is written entirely on your device with an open-source spreadsheet library. Your financial or business data is never uploaded β the only network request is the one-time download of the library.