Excel changed my CSV dates. Here's how to fix it.

You had a clean column of dates like 03/04/2026. You opened the file in Excel, saved it, and now that same cell says 4-Mar-26. Or worse, it shows 44655 — a random-looking five-digit number. The dates you carefully typed are gone, and the file you need to send out is wrong.

before: 03/04/2026 after: 4-Mar-26 (or a serial number like 44655)

Why Excel does this

Excel doesn't leave CSV files alone. The moment you double-click a .csv, Excel reads every column and guesses what kind of data it holds. When it sees something that looks like a date, it converts the text into its own internal date value and then shows it using whatever date format your computer is set to. That's where 4-Mar-26 comes from.

The 44655 is the same thing one step further. That's Excel's internal "day number" (days counted from 1900) showing through because the cell ended up formatted as General or Number instead of Date. Either way, once you save, the original text 03/04/2026 is no longer in the file.

The manual fix: import, don't open

Don't double-click the CSV. Bring it in through the import tool so you can tell Excel to keep the column as plain text:

1. Open Excel to a blank workbook first.
2. Go to Data ▸ From Text/CSV (older Excel: Data ▸ Get External Data ▸ From Text).
3. Pick your file. In the preview window, click the date column's header and set its type to Text (choose "Text" in the dropdown, or select the column and click the "Text" radio button in the legacy wizard).
4. Load it. Excel now keeps 03/04/2026 exactly as it was.

If the damage is already saved, you can sometimes reformat: select the column and set the cell format back to a date format you like. That only works while the values are still real dates — not if they've already been flattened to text like 4-Mar-26.

← Back to the CSV Fixer tool