How to open a bank CSV in Excel without breaking dates

Double-clicking a CSV opens it in Excel with your computer’s regional settings. With bank data that causes three classic problems:

• Dates flip or stay as text. A file with 04/05/2025 meaning 4 May can be read as April 5 on a US-configured computer, and dates that do not fit your settings stay as text that will not sort.
• Long reference numbers break. Excel keeps only 15 significant digits, so a 16- or 18-digit reference becomes something like 1.00000E+17 and its last digits are lost.
• Leading zeros disappear from account numbers and sort codes.

The fix is to import the CSV instead of opening it, and tell Excel what each column is.

In Excel for Microsoft 365 (Power Query)

1. Open a blank workbook and choose Data → From Text/CSV, then pick the file.
2. In the preview, set File Origin to 65001: Unicode (UTF-8) if accented names look wrong, and check the delimiter (comma or semicolon).
3. Click Transform Data.
4. For the date column: right-click the header → Change Type → Using Locale… → Data type Date, and pick the locale the file was written in, for example English (United Kingdom) for 31/01/2025 or English (United States) for 01/31/2025.
5. For reference or account number columns: Change Type → Text, so no digits are lost.
6. For amounts: Change Type → Decimal Number (use “Using Locale…” again if the file uses a comma for decimals).
7. Close & Load.

In older Excel versions

Use Data → Get External Data → From Text (the Text Import Wizard). In step 3, select the date column and choose Date with the right order (DMY or MDY), and set reference columns to Text.

In Google Sheets

Use File → Import → Upload. Sheets reads dates according to the spreadsheet’s locale (File → Settings → Locale), so set the locale that matches the file before importing.

Avoid the problem entirely

CSV files from StatementEasy use ISO dates (2025-01-31) by default, which Excel and Sheets read the same way everywhere, and amounts with a dot for decimals. If you just want to work in Excel, download the .xlsx instead: dates and amounts are already real values, nothing to import.