Pages
Choose all, odd, even or specific table-bearing pages.
Including pages with prose or repeated headers can add unwanted rows to the CSV.PDF to CSV
Export rows and columns estimated from PDF text into a CSV file. Open the result in a spreadsheet or use it as a starting point for data cleanup.
Drop your PDF file here
Choose a file to get started.
Files stay on your device No sign-up Free to use
How it works
Select pages containing a table
Select a PDF containing text-based tables.
Extract the estimated rows
Choose the pages that contain the data.
Verify cells before spreadsheet use
Download the CSV and verify row and column alignment.
PDF to CSV
Convert from PDF
Extract text-based tables into CSV for spreadsheets
Move a price list into a spreadsheet
A short price table needs spreadsheet cleanup. Export its page, preserve the original and review every amount before calculations. Extract only pages containing the table. Open the CSV as text first to inspect separators, then correct column alignment before numerical calculations.
✓ Check row count, column alignment, decimal separators and leading zeros; compare totals with the source.Make it work for your document
Options after selecting a fileFrequently asked questions
The details, when you need them.
Can this recover formulas?
No. It exports visible values and text only.
Why do some text cells start with an apostrophe?
It keeps formula-like text from being interpreted as a spreadsheet formula.
Best for regularly aligned tables
Columns are estimated from where text appears on the page. Consistent spacing and straightforward rows produce better results than merged cells, wrapped descriptions or overlapping tables. The export cannot recover the formulas or spreadsheet formatting used to create the PDF.
Validate the extracted data
Pages are separated by blank rows in the CSV. Inspect totals, identifiers and decimal formatting before using the data. Text that begins like a spreadsheet formula receives a protective apostrophe. Numeric conversion may remove leading zeros from identifiers, so compare the output with the source. Image-only tables need OCR first.
Estimated columns require cell verification
Column inference is approximate. Merged cells, wrapped descriptions and repeated page headers can produce incorrect rows.
Amounts appear in the wrong columns
The converter estimates columns from text positions, which can be ambiguous. Compare every financial row with the source and correct misaligned cells before formulas or imports.