Why PDF tables are hard to copy
A PDF stores text as positioned pieces, not as rows and columns. That's why copying a table out of a PDF and pasting it into Excel usually lands everything in one column, or splits one cell across three. Scanned PDFs are worse, because there's no text at all, only a picture of the page.
This tool reads the page the way a person does. It finds each table, works out where the rows and columns are, and rebuilds it as a real spreadsheet. It handles both kinds of PDF, the ones made by software and the ones made by a scanner.
You also don't have to stitch tables back together. When a table continues onto the next page with the same columns, the parts are joined into one table and the repeated header row is dropped.
What you get in the workbook
One .xlsx file per PDF, with filters already switched on for every header row.
| Part of the PDF | In Excel |
|---|---|
| Each separate table | Its own sheet, named after the table title when there is one. |
| A table across several pages | One continuous table on one sheet. |
| Numbers | Real numbers where possible, so SUM and pivot tables work straight away. |
| Merged cells | The value is repeated in each covered cell, which makes sorting and filtering behave. |
| Two-level headers | Flattened into a single header row, for example "2025 Revenue", "2025 Cost". |
| Page headers, footers, body text | Left out. Only table data is kept. |
How to convert a PDF to Excel
Upload the PDF
Drop in up to 10 PDFs at once, each up to 10 MB. Images of tables work too.
Convert
Each PDF is read page by page. Short documents take a few seconds; longer ones take a little more.
Review the tables
Every table appears as a tab in the preview. Check the totals or a few rows against the PDF.
Download
Get the whole workbook as .xlsx, or the table you are viewing as CSV.
What people convert
Financial reports
Annual reports and management accounts with balance sheets and P&L tables you want to model.
Supplier price lists and catalogues
Long product tables with codes, descriptions and prices across many pages.
Government and research data
Statistical tables that are only published as PDF.
Scanned paperwork
Old printed records that were scanned to PDF and never typed up.
Tips for better results
- Documents up to about 15 to 20 pages work best in one go. For a long report, use Split PDF to keep only the pages with the tables you need.
- For scans, 200 to 300 DPI is plenty. Very low-resolution scans lose small digits.
- If a page is sideways, rotate it first with Rotate PDF. Upright pages are read more reliably.
- Compare one column total with the PDF. If it matches, the rest of the column almost certainly does too.
What to watch for
- Password-protected PDFs need to be unlocked before uploading.
- Very dense tables with tiny print can have a few misread digits. Zooming the scan or splitting the page helps.
- Charts and diagrams aren’t converted, only tables of text and numbers.
- Cell colours, borders and fonts are not copied. You get the data, not the styling.
Frequently asked questions
Does it work with scanned PDFs?
Yes. Scanned PDFs are read with AI text recognition, the same way as digital ones. Clear, upright scans give the best results.
What if one table spans several pages?
If the table continues with the same columns, the pages are joined into one table and the repeated header rows are removed.
How are several tables in one PDF handled?
Each table is put on its own sheet in the Excel file. In the preview, you can switch between them with the tabs and download any one as CSV.
Is there a page limit?
The file limit is 10 MB. In practice, documents of up to about 15 to 20 pages convert reliably in one go. Split longer PDFs into parts first.
Will the numbers work in formulas?
Yes. Values that are plain numbers are saved as numbers, so SUM, AVERAGE and pivot tables work without fixing anything. Codes with leading zeros are kept as text.
Is it free?
You get free daily credits, and each file uses one credit. If you need more, there are optional one-time credit packs. There is no subscription.
Is my PDF stored anywhere?
Your file is sent to Google's Gemini AI to be read, and the result comes straight back to your browser. Toolghar doesn't save the file or the extracted data, and the spreadsheet is built on your device.