PDF to Excel Converter: Free PDF to XLSX with OCR

It is a free online tool from Entertainment Couch that turns PDF files into editable Excel spreadsheets (.xlsx) in your web browser. It finds the tables on each page, rebuilds their rows, columns and merged cells, and turns numbers and dates into real Excel values. Scanned PDFs and photos of tables are read with OCR. There is no sign-up, no watermark and nothing is uploaded.

Yes. Every feature is free with no daily limit, no page quota, no premium plan and no watermark. You can convert as many PDFs as you like.

No. Open the page, add your PDF and convert it. There is no account, email, app or browser extension.

There is no page limit. You can add up to 50 files at a time, each up to 1 GB. Because conversion runs on your device, very large scanned PDFs depend on your device’s speed and memory. Digital PDFs convert in well under a second per page; scanned pages usually take about 5 to 15 seconds each, depending on your device and the OCR quality setting.

No. Your Excel file contains only the data from your PDF.

Press Choose PDF files or drag your PDF onto the tool, then press Convert to Excel. Check the tables in the preview, edit any cell if you need to, and press Download. The whole process usually takes less than a minute.

Yes. Add up to 50 PDFs or images together and press Convert to Excel once. Each file becomes its own Excel workbook, and All (ZIP) downloads every workbook in one ZIP file.

Yes. Under Pages choose Choose pages and type the pages you want, such as 1-3, 5, 8-10. Only those pages are read, which also makes large files faster.

Yes. Under Excel sheets choose One per page, One per table or All in one. One per table also adds a bold header row, filter buttons and a frozen top row. All in one with Tables only joins a table that runs across several pages into one long table.

Yes. After converting, the CSV button downloads the sheet you are viewing as a UTF-8 CSV file. If the PDF uses commas as decimal separators, the CSV uses semicolons between columns so it opens correctly in European versions of Excel.

Yes. Press Copy after converting, then paste into Google Sheets, Excel or Numbers. You can also upload the downloaded .xlsx file to Google Drive and open it with Google Sheets.

Yes. The Excel data view shows every sheet as a grid. Click any cell to change it, insert or delete rows, delete columns, or turn the header row on or off. Your edits are saved into the Excel file you download.

Yes. Add JPG, PNG, WebP, GIF or BMP images, or paste a screenshot. Images are always read with OCR, and the table is rebuilt the same way as in a PDF. On a phone, Take a photo lets you photograph a printed table directly.

Yes. With Text recognition set to Auto, the converter checks every page. Pages that have real text are read directly, and scanned pages are read with OCR, then their tables are rebuilt as rows and columns. You do not need to do anything different for scanned files.

OCR (optical character recognition) reads text from an image of a page. The converter uses it for scanned PDFs, photos and pages whose text cannot be copied. Auto decides per page; Every page forces OCR on all pages, and Off never uses it on PDFs.

English, Arabic, Bengali, Chinese (Simplified), Dutch, French, German, Hindi, Italian, Japanese, Korean, Persian (Farsi), Portuguese, Russian, Spanish, Turkish and Urdu. You can pick a second language for documents that mix two, such as English and Urdu.

Only to a limited degree. Neat, separated block capitals and printed-style digits are often read correctly, but joined-up cursive handwriting usually is not. The OCR engine is trained on printed text, so check handwritten results carefully in the Excel data view before downloading.

On clean scans at 300 dpi, printed text is usually read almost perfectly. Accuracy drops with blur, low resolution, heavy shadows, stamps over text and decorative fonts. The preview shows an OCR confidence score for each page, and you can fix any cell before you download.

With Straighten and auto-rotate scans turned on (the default), the converter straightens pages tilted by up to about 6 degrees. If the text on a page reads badly, it also tries the page turned 90, 180 and 270 degrees and keeps the best result.

Reading text from an image is much more work than copying text that is already in the PDF. Each scanned page is rendered as a high-resolution image, cleaned up, checked for table lines and then read word by word, all on your device. Choose OCR quality › Fast to speed it up.

It uses two methods. For bordered tables it reads the lines drawn in the PDF (or detected in a scan) and builds the grid from them, including merged cells. For tables without borders it groups words into lines and finds columns from how the text is aligned down the page. Text between tables is kept as normal rows.

Yes. In bordered tables a cell that spans several rows or columns becomes a merged cell in Excel. In tables without borders, a heading centred over several columns is also merged across them. You can turn this off under More options › Rebuild merged cells.

With Content set to Tables and text, headings, notes and other text are kept in order above and below the tables. Choose Tables only to keep just the tables.

In the page preview, press Draw table area and drag a box around the table: that area becomes one table. Click a table outline to leave it out. If columns run together, set Column detection to More columns; if one column is split in two, choose Fewer columns.

Yes. With Join tables that continue on the next page turned on, a table at the top of a page that has the same columns as the table at the end of the previous page is added to it, and a repeated header row is removed. This works with One per table and All in one.

No. The converter extracts data, not design. It keeps bold header rows where it can detect them and adds its own clean header styling, but it does not copy fonts, colours, logos, images or charts from the PDF.

Yes. Values such as 1,234.56, $1,200, 15%, (2,500.00) and 1,500.00 CR become real numbers you can add up, with a number format that looks like the original. Negative numbers in brackets or with a trailing minus become negative values.

Yes. Number format › Detect works out from the whole document whether commas or points are the decimal separator. You can also choose 1,234.56 or 1.234,56 yourself.

Dates such as 22/10/2026, 2026-10-22, 22 Oct 2026 and Oct 22, 2026 become real Excel dates. For dates like 03/04/2026 the converter checks the whole column: if any value only makes sense one way round it uses that, otherwise it follows the Dates setting (Day first or Month first). Choose Keep as text to leave dates exactly as written.

Values that are really identifiers stay as text so nothing is lost: numbers with leading zeros (00123), numbers longer than 15 digits such as card and account numbers (Excel only stores 15 significant digits), and phone numbers that start with +.

No. A PDF only stores the printed results, so totals and subtotals come through as values, not formulas. You can add your own formulas in Excel after downloading.

No. The converter runs entirely in your browser. Your PDF is read, converted and saved as an Excel file on your own device and is never sent to Entertainment Couch or anyone else.

Yes, because nothing leaves your device. You can check this yourself: load the page, turn off your internet connection, then convert a digital PDF. It still works.

Yes, if you know the password. You will be asked for it when you add the file. The password is used only on your device to open the PDF and is never stored or sent. The converter cannot remove a password you do not know.

Once the page has loaded, digital PDFs convert without a connection. Scanned PDFs need the OCR engine, which downloads the first time you use OCR in a language; after that, the language file is kept in your browser’s storage.

Yes. The converter works in modern mobile browsers on Android, iPhone and iPad. On a phone you can also photograph a printed table with Take a photo. Very large scanned PDFs are faster on a computer.

Current versions of Chrome, Edge, Firefox, Safari and Samsung Internet on Windows, macOS, Linux, Android, iPhone and iPad. JavaScript must be turned on.

Yes. The download is a standard .xlsx file in the Office Open XML format, which Microsoft Excel, Google Sheets, Apple Numbers and LibreOffice Calc can all open.

Some PDFs contain text with a broken font encoding, so copying it produces gibberish. Set Text recognition to Every page and convert again: the pages are then read as images with OCR. If the file is a scan, check that the right Document language is selected.

Change More options › Column detection: More columns splits values that ran together, Fewer columns joins a column that was cut in two. The preview updates straight away. For one difficult table, use Draw table area.

Convert fewer pages at a time with Pages › Choose pages, set OCR quality to Fast, and close other heavy browser tabs. Scanned PDFs with hundreds of pages are best converted on a computer.

The file may be incomplete or not really a PDF. Download it again from where you got it, or open it in your PDF reader and use Print › Save as PDF to make a fresh copy, then convert that.