Can it extract tables from scanned PDFs?

Not directly. Each page of a scan is an image with no text positions to work with, and the tool tells you when it detects one. Run the file through our Scanned PDF OCR tool first to add a text layer, then extract the tables.

The columns are split in the wrong place. What can I do?

Drag the blue lines in the “Adjust column splits” diagram to move a split, double-click an empty spot to add one and click the × at the top of a line to remove it; the preview updates immediately. If a table was missed entirely, switch to “Treat each page as one table” and adjust from there.

What happens to merged cells and tables that span pages?

Merged cells are split and their text goes into the cell at its position. A table that continues onto the next page becomes one table per page; choose “All tables in one sheet” to stack them together. Fonts, colours and borders are not kept.

Are my files uploaded?

No. Parsing the PDF, detecting tables and building the spreadsheet all happen in your browser with no network upload, so it is safe for statements and other sensitive data.

📗

📗PDF to Excel - Extract Tables from PDF to XLSX or CSV

Financial statements, bank statements, price lists and grade reports often arrive only as PDFs, and retyping every number into a spreadsheet is slow and error-prone. This tool reads the position of every piece of text in the PDF and rebuilds the tables as rows and columns. Ruled tables are split along the horizontal and vertical lines drawn on the page, keeping line breaks inside cells; tables without borders are found by grouping text into rows by baseline and looking for vertical white-space gaps shared by those rows, while ordinary paragraphs and two-column body text are skipped. Detected tables are listed as “Page · Table”, and you can preview each one and leave out the ones you don’t need. If a column is split in the wrong place, drag the blue split line on the diagram, double-click to add a split or click × to remove one, and the grid preview updates instantly; if a table is missed, switch to “Treat each page as one table”. On export, text such as 1,234.56, -12, (300) and 12% becomes real numbers you can sum and sort (with thousands and percent formats kept), while IDs with leading zeros and very long numbers stay text so nothing is lost. Choose one sheet per table, one per page or everything in one sheet, or export UTF-8 CSV that opens cleanly in Excel. Everything runs locally in your browser; the file is never uploaded.

Free · No sign-up · Runs in your browserLast updated: 2026-10-26

Loading tool...

How to use PDF to Excel

1Upload a PDF that contains tables (a text-based PDF; run scanned files through OCR first)
2Review the detected tables, untick any you don’t need, drag column splits to fix the layout if necessary, or switch to treating each page as one table
3Pick Excel or CSV, how to organise sheets and whether to convert numbers, then click Export tables

FAQ

Can it extract tables from scanned PDFs?: Not directly. Each page of a scan is an image with no text positions to work with, and the tool tells you when it detects one. Run the file through our Scanned PDF OCR tool first to add a text layer, then extract the tables.
The columns are split in the wrong place. What can I do?: Drag the blue lines in the “Adjust column splits” diagram to move a split, double-click an empty spot to add one and click the × at the top of a line to remove it; the preview updates immediately. If a table was missed entirely, switch to “Treat each page as one table” and adjust from there.
What happens to merged cells and tables that span pages?: Merged cells are split and their text goes into the cell at its position. A table that continues onto the next page becomes one table per page; choose “All tables in one sheet” to stack them together. Fonts, colours and borders are not kept.
Are my files uploaded?: No. Parsing the PDF, detecting tables and building the spreadsheet all happen in your browser with no network upload, so it is safe for statements and other sensitive data.