PDF to CSV

Extract tables from a PDF with selectable text into CSV files, one file per table. Scanned pages need OCR first. Files stay in your browser.

Small converter download on first use.

Detected tables

Was this tool useful?

Try an example

Download sample

The CSV should contain four rows and three columns, with Item, Qty, Amount and the total 148.50. Multiple tables download as separate CSV files in a ZIP.

What to expect

For PDFs with embedded text and regular table rows. Scans need OCR first. Detection can miss tables or place text in the wrong cell; compare the downloaded file with your PDF. Merged cells and original formatting are not reconstructed.

Each table becomes its own CSV file or Excel sheet. Values remain text. CSV values that could be interpreted as formulas are prefixed with an apostrophe; Excel cells are explicitly stored as text. When importing CSV into a spreadsheet, choose text columns to preserve leading zeros and number formatting.

Up to 10 MiB input, 200 pages, 100 tables, 100,000 cells, 256 columns and 8 MiB output. Processing stops after 30 seconds. Irregular rows are rejected rather than silently shifted.

Need raw extraction?PDF tables to JSON.

The first run may download converter files and take a moment. If it fails, check your connection or try a smaller file. Cancel stops active processing; you can choose another file and retry.

Engine licenses and source code

A practical guide

Table detected, CSV created.

Choose a PDF with selectable text.
Run PDF to CSV.
Check each CSV's rows and columns against the source table.

Try this example

Extract a table with three columns and eight rows from a PDF; the CSV should contain those same eight rows and three columns.

Before you start: Detection can miss complex or borderless tables, and scanned pages produce no result until they go through OCR first.