The short answer
PDF to Excel finds the columns from where the text sits and writes real spreadsheet rows, with amounts as numbers that add up. For a table that runs across pages, keep "All in one sheet": the repeated header is dropped and the rows join into one table.
Runs in this tab, on pdf.js, Tesseract (WebAssembly) and ExcelJS, so your file is read from your own device and never sent anywhere. Check it yourself
Step by step
- Make sure the PDF has text, not a picture Select a number on the page. If nothing highlights, it is a scan and there are no columns to read yet.
- Convert with the default layout "All in one sheet" suits statements and long reports. Use one sheet per page when each page is a separate table.
- Check the first and last rows against the PDF That catches a table that started a row early or late, which is what a page range fixes.
- Total a column Sum the amounts and compare with the document’s own total. Matching means the parse is right.
- Use CSV if something else will read it PDF to CSV gives the same table as plain text for a script, a database import or an accounting package.
Questions
Do the amounts come across as numbers?
Yes, with their formatting kept — currency, separators, decimals — and negatives in brackets or marked DR stay negative. Long codes stay text so they keep leading zeros.
What about a description that wraps onto two lines?
It is rejoined to its row rather than becoming an empty row, which is why the row count matches the number of transactions.
Is my statement uploaded?
No. The PDF is read and the workbook written in your browser.
More in PDF to Word, Excel & PowerPoint .