How to convert a PDF table to Excel for free, without signing up
You have a table stuck in a PDF and you need it in a spreadsheet. Copy and paste mangles it, and most converters want an email address first. Here is a way that does not, and what to expect from the result.
Last checked 2026-11-06.
Why copy and paste goes wrong
A PDF does not contain tables the way a spreadsheet does. It contains pieces of text, each placed at a position on the page. What looks like a neat grid on screen is just words that happen to line up. When you copy a selection, your PDF reader guesses at the order, and usually puts every value on its own line or runs the columns together. That is why a ten-column table so often pastes into Excel as one long column.
A converter has to rebuild the table: group the words into rows by how far down the page they sit, then into columns by where they start across the page. That is what the PDF to Excel tool on this site does.
Converting, step by step
- Drop in your PDF. There is no account, no email and no signup. The free tier handles files up to 10 MB, with a daily cap per tool.
- Choose the output. An Excel workbook (.xlsx) puts each PDF page on its own sheet, named Page 1, Page 2 and so on. A CSV puts every page into one file, which is handy for importing into accounting software or a database.
- Leave "Read numbers as numbers" on (it is on by default) so the spreadsheet can add up amounts straight away instead of treating them as text.
- Check the preview. The page shows how many rows it found and the first few rows, before you download.
- Download and open in Excel, Google Sheets, Numbers or LibreOffice.
If your PDF is a scan
Some PDFs are pictures of pages, made by a scanner, a phone camera or a fax. They have no text inside to read, which is why many converters return an empty file. Try selecting a word in your PDF reader: if you cannot highlight it, it is a scan.
When the tool finds no text, it offers to read the page with OCR (optical character recognition). The OCR engine, Tesseract compiled to WebAssembly, is downloaded once, about 6 MB, and then runs inside the page on your machine. Most other sites that offer OCR upload your document to their servers to do it.
OCR is slower than reading a text PDF and less exact. On a clean, straight scan made at 200 to 300 dpi it usually reads well, and the tool tells you the confidence it measured so you know how carefully to check. Faint photocopies, handwriting and photos taken at an angle read much worse. If OCR finds nothing, a sharper scan made at 300 dpi usually fixes it. There is a full guide to scanned PDFs.
What you get, and what you do not
You get the data: rows, columns, and numbers as numbers, ready to sort, filter and total. You do not get the look of the original: fonts, colors, borders and merged cells are not carried across. For most jobs that is what you want, a clean table rather than a picture of one.
Regular tables, with one line per row and values that line up in columns, come out clean. Tables where a cell wraps onto two or three lines, or where columns are not aligned, come out rougher and need tidying.
Fixing a messy result
Most problems fall into a few patterns, and each has a quick fix in the spreadsheet:
- A value landed in the wrong column. This happens when a column is ragged, for example a description that is sometimes long enough to push into the next column. Sort by the problem column to group the odd rows, then move them in one go. The wrong-column guide covers this in detail.
- One table row became two spreadsheet rows. A cell that wraps onto a second line becomes its own row. Join the text back with a formula such as `=A2&" "&A3`, then delete the extra row.
- Headers and footers came along. Page titles, page numbers and "continued" lines appear as rows. Filter them out, or delete them before totalling.
- Numbers came through as text. Check that "Read numbers as numbers" was on. Currency symbols and thousands separators in unusual formats can also confuse it; Excel's Text to Columns or Find and Replace sorts them out.
Bank statements, invoices and other sensitive tables
The tables people most often need out of a PDF are financial ones: bank and card statements, invoices, payslips, price lists. Those are exactly the documents you should think twice about uploading to a stranger's server.
This tool never uploads them. Text reading uses pdf.js and the OCR uses Tesseract, both running in the page. You can check: open your browser's developer tools, choose the Network tab, and convert a file. No request carries your PDF. There is a guide for bank statements specifically, including how to handle running balances and multi-line transactions. For a whole folder of statements, see bank statements in bulk.
Limits worth knowing
- Password-protected PDFs are not supported. Remove the password first, then convert.
- Free use has a daily cap per tool and a 10 MB file limit. Plans that remove the cap and accept files up to 200 MB are described on Free vs Pro; paid checkout is not open yet.
- Very large scanned files take a while, because OCR runs on your own processor.
Use something else when you need pixel-perfect layout recreation, or when a bank already offers a CSV or OFX export: a real export beats any conversion. This tool is for the PDF-only case.
Questions people ask
Can I convert a PDF table to Excel without signing up?
Yes. There is no account, email or signup on this site at all.
Does it work on scanned PDFs?
Yes. It offers OCR when the PDF has no text, and the OCR runs on your computer.
Excel or CSV?
Excel gives one sheet per page. CSV gives one file with every page, which is easier to import elsewhere.
Will it keep my formatting?
It keeps the structure and numbers, not fonts, colors, borders or merged cells.
Is my PDF uploaded?
No. Nothing is sent; you can check in the Network tab.
Does it work with password-protected PDFs?
No. Remove the password first, then convert that copy.
Try it
The conversion runs in this page, with OCR on your own machine for scans, so the PDF never leaves your computer.