Convert a bank statement to CSV

What is a CSV bank statement?

A CSV (comma-separated values) bank statement is a plain-text file with one transaction per line and the values of each line separated by commas. Because it's just text, almost every program can read it: Excel, Google Sheets, Numbers, budgeting apps and accounting software.

Most banks let you download recent transactions as CSV from online banking, but each bank uses its own layout: some split debits and credits into two columns, some leave out the running balance, some add header lines above the data. Older periods are often only available as PDF statements. Converting the PDF gives you a CSV for any period, in the same layout for every bank.

Bank statement CSV format: the columns you get

The columns are the same whatever bank or language the statement came from, so a file from one bank can be appended to a file from another.

Amounts have no currency symbols or thousands separators, so they import as numbers instead of text.

Comma or semicolon: making the CSV open correctly in Excel

Excel splits a CSV into columns using the list separator from your computer's regional settings. In the US, UK and many other countries that's a comma and amounts use a decimal point (1234.56). In much of Europe and Latin America Excel expects a semicolon between values and a decimal comma (1234,56) — open a comma file there and everything lands in one column.

So the export has two formats. Comma writes 1234.56 with commas between values; semicolon writes 1234,56 with semicolons between values. Pick the one that matches the program and the region that will open the file. If a file still opens in a single column, use Data → From Text/CSV in Excel and choose the delimiter yourself.

The file is saved as UTF-8 with a byte-order mark, which tells Excel the encoding. Names like Müller, Café or Łódź come through intact instead of turning into garbled characters.

Date formats

Choose the date format the receiving program expects. Getting it wrong is the most common reason an import fails or puts transactions in the wrong month.

• YYYY-MM-DD (2026-03-04) — unambiguous and sorts correctly even as text; the safest choice for databases and scripts
• MM/DD/YYYY (03/04/2026) — US software
• DD/MM/YYYY (04/03/2026) — UK, Australia, India and most of the world
• DD.MM.YYYY (04.03.2026) — Germany, Austria, Switzerland, Poland and others

03/04/2026 means 4 March in the US and 3 April almost everywhere else. If dates look shuffled after an import, the date format is the cause.

Safe to open: no formula surprises

Spreadsheets treat a cell that starts with =, +, - or @ as a formula. A transaction description that happens to start with one of those characters could be run as a formula when you open the CSV — a known risk called CSV or formula injection.

Text cells that start with those characters are written with a leading apostrophe, so they open as plain text. Amounts are written as plain numbers and stay numbers, so negative values still work in SUM and pivot tables.

Importing the CSV into other apps

• Google Sheets: File → Import → Upload, then choose comma or semicolon as the separator.
• Excel and Numbers: open the file directly, or use the import options above if the columns don't split.
• Budgeting apps: most accept a CSV with a date, a description and one signed amount column — which is exactly what this file has.
• Accounting software: use the export made for it. QuickBooks Online and Xero expect their own column layouts, so export the QuickBooks CSV or the Xero CSV instead of the plain CSV.
• Apps that prefer bank files: export OFX or QIF.

Already have your bank's CSV?

Upload it anyway. Your bank's CSV, Excel, OFX or QIF downloads are read and categorized the same way as PDFs, and they use no pages (you still need a plan or page pack). You get the same clean columns, payee names and categories as for your PDFs, so you can mix PDF months and CSV months in one export.

Select several statements and export them together to get one CSV in date order, optionally limited to a date range. If you want totals by category and month as well, export to Excel instead: the workbook adds a Summary sheet.

Questions