Pull tables out of a PDF into an Excel spreadsheet, with numbers stored as real numbers you can calculate with. The PDF is read on your device and never uploaded, so it's suited to bank statements and payslips, with no page limit or sign-up.
How to use PDF to Excel
Choose a PDF to convert to Excel.
Under Sheets, choose One per page or All tables in one.
Preview the detected tables in the tabs and check the columns.
Click Download .xlsx, or Download CSV for the current sheet.
Frequently asked questions
How do I convert a PDF table to Excel?
Open the PDF here. Tables are detected from how the text lines up into columns and shown as a preview; click Download .xlsx to save them as a spreadsheet.
Are numbers kept as numbers?
Yes. Values like 1,234.50, (1,200), 12.5% and ₹1,00,000 (Indian digit grouping) are stored as numbers, so sums and formulas work straight away. Codes with leading zeros and long digit runs such as account numbers stay as text, so they aren't mangled.
Why does one row of my table become two?
When a cell's text wraps onto a second line in the PDF, that line is read as an extra row. Merge those rows by hand in Excel after converting.
Can I convert a bank statement PDF to Excel?
Yes, if it's a text PDF downloaded from net banking rather than a scan. Statements are often password-protected; remove the password first with Unlock PDF, using the password your bank gave you, then convert it here.
What if a page has no table?
With one sheet per page, its text lines are listed in a single column instead, so nothing is lost.
Can I convert a scanned PDF to Excel?
Scanned pages have no text to read. Run OCR PDF on the file first, then convert it here.
Is my PDF uploaded?
No. Tables are found and the spreadsheet is built in your browser.
What converts well
Need the text rather than a spreadsheet? Use PDF to Word.
Converts well: statements, invoices, price lists and reports exported from software, with clear columns separated by space.
Needs checking: tables where cells wrap onto two lines, which become extra rows.
Converts poorly: scans without OCR, and two-column text that isn't a table, which can be mistaken for one.