RunToolRunPDF to Excel
Convert PDF files to Excel format.
How to Convert PDF to Excel
Select a PDF with tables
Click the upload area or drop a text-based PDF containing tabular data.
Rows are reconstructed locally
Text positions are analyzed to rebuild rows and columns, entirely in your browser.
Download the .xlsx file
Open the result in Excel, Google Sheets or LibreOffice and continue working with the data.
A table in a PDF is not a table
PDFs store text as glyphs at coordinates. What looks like a table is text positioned in a grid, sometimes with lines drawn around it and sometimes not. There are no cells, no rows and no columns in the file — those are things your eye infers from alignment.
Converting therefore means reconstructing structure from position: grouping text into rows by vertical alignment and into columns by horizontal gaps. It works well on clean, regular tables and struggles exactly where a human would need a moment to work out what belongs where.
What breaks the reconstruction
Merged cells, multi-line entries inside a single cell, and tables that continue across a page break with the header repeated are the common failures. Numbers formatted with thousands separators or currency symbols may arrive as text rather than as values, which quietly breaks every formula you then write.
Financial statements are a particular trap: negative numbers shown in parentheses, footnote markers attached to figures, and totals separated by rules rather than by blank rows. Always check the totals in the converted sheet against the PDF before using the data.
Scanned documents will produce nothing
If the PDF came from a scanner, its pages are images and there is no text to extract. The conversion will yield an empty or near-empty spreadsheet, and no setting will change that. The two-second test is to try selecting a number in the PDF with your cursor: if it highlights, extraction will work.
Recovering a table from a scan requires optical character recognition, which reads characters from pixels and makes mistakes — particularly with digits, where a misread 8 for a 3 is both easy and expensive.
Check before you calculate
Treat the output as a draft that needs verification, not as data. Compare a few rows against the source, check that column totals match the printed totals, and confirm that numeric columns are actually numeric rather than text that merely looks like it.
Where the underlying data exists somewhere else — a report system, a database export, a colleague's original spreadsheet — that source is always better than reconstructing it from a document designed for printing.
Why it runs in your browser
Financial statements, invoices and internal reports are the overwhelming majority of what gets converted this way, and they are exactly the files that should not be uploaded to an unfamiliar service. Extraction and spreadsheet generation both happen on your device.
The practical limit is document length rather than policy: pages are processed one at a time, so a long report takes proportionally longer, and converting the specific pages you need is faster and easier to check than converting all of them.