The cleanest way to get a statement into CSV is to download it from the bank in that format, if the bank offers it for your account. If it only gives a PDF, you can copy the table into a spreadsheet, tidy the columns and save as CSV. Either way, finish by checking that your total matches the statement, because a converted file is only useful if it is complete. The steps below are the general method, and the exact buttons differ by bank, so check your own statement page.
Does your bank already offer a CSV or Excel download?
Look on the statement or account activity page for a download or export option listing file types. Some accounts offer CSV for transaction history even when the monthly statement is a PDF. Choose the exact date range of the statement and keep the file name clear, such as card-2026-03.csv, so you can find it later.
How do you convert a PDF when there is no CSV option?
- Open the PDF on your computer and select the transaction table with the mouse.
- Copy it and paste into a blank Excel or Google Sheets sheet.
- If everything lands in one column, use Text to Columns in Excel, or Split text to columns in Google Sheets, with space or tab as the separator.
- Delete headers, footers and page-break rows so each row is one transaction.
- Keep four columns: date, description, amount, and a note column for anything you add.
- Save as CSV from the File menu.
Scanned statements are images, so copying gives nothing. In that case ask the bank for a normal download, because retyping hundreds of rows invites mistakes.
How do you fix dates and signs?
US-style dates read month first, so 03/04/2026 means March 4 and not 3 April. If a spreadsheet in India reads it the other way, you get wrong months. Fix it before saving. Many statements also show spending as negative and payments as positive. If you prefer spending as positive numbers, add a helper column with this formula in the first data row and fill down: =-C2. For text dates such as 03/04/2026 sitting in column A, this formula builds a true date: =DATE(RIGHT(A2,4),LEFT(A2,2),MID(A2,4,2)). Format the result as a date and keep the original column until you are sure.
How do you check that the CSV matches the statement?
Say your converted sheet has these spending rows in dollars: 42.10, 18.75, 120.00 and 9.15. Add them: 42.10 plus 18.75 is 60.85, plus 120.00 is 180.85, plus 9.15 is 190.00. Then compare with the statement's total for debits, which should read 190.00. In Excel, =SUM(C2:C5) does the same job. If the numbers differ, look for a missing row at a page break, a fee line you cut, or a refund that flipped the sign.
| Check | How | What a mismatch usually means |
|---|---|---|
| Row count | Count transactions on the statement and in the sheet | A page-break row was deleted |
| Total debits | =SUMIF(C2:C500,"<0") for negative spends | A refund or fee changed sign |
| Date range | =MIN(A2:A500) and =MAX(A2:A500) after fixing dates | A date read in the wrong order |
What then? Importing it for tracking
SpendixAI takes bank statements as PDF, CSV or Excel, so a tidy CSV works, and so does the original file. You pick the account first and review the rows before saving. For dollar statements, choose your base currency before the first import, since it cannot be changed after transactions exist, and each expense keeps its own currency with the rate saved against it, so totals stay readable.
Useful ideas start with a clearer picture of your own spending.
Explore your reports

