What Garbled Characters Mean
You download a bank CSV and open it in Excel, but a merchant called "Café Lumière" appears with odd symbols in place of the accents, or a pound or euro sign is replaced by unrelated characters. The characters in the file are usually fine. The problem is that a CSV file is plain text, and Excel has to decide which character set to use to read it. If it picks a different one from the one used to write the file, the letters come out wrong.
Microsoft's documentation for the Text Import Wizard describes the setting that controls this, called File origin, as the character set that is used in the text file. The page says that in most cases you can leave it at the default, and that you should change it when you know the file was created with a different character set from the one your computer uses.
Why Double-Clicking Gives You No Control
Microsoft's help for importing text files says that when you open a .csv file in Excel it opens directly in a new workbook, and that Excel applies its current default data format settings to the columns. That means no import dialog appears in which you could choose a character set. To get a choice, import the file instead of double-clicking it.
Import the CSV and Set File Origin
Microsoft documents an import route for newer versions of Excel. On the Data tab, select Get Data, then From File, then From Text/CSV. Browse to the CSV file. Microsoft's Azure Cost Management guide, which addresses exactly this problem for garbled exports, then says to set File origin to 65001: Unicode (UTF-8) and select Load. Microsoft's Partner Center guide gives matching steps and adds that you can alternatively use a specific Unicode code page identifier for your language, and that you should set the Delimiter to Comma.
In summary, the steps those pages list are:
1. Data tab > Get Data > From File > From Text/CSV
2. Select the CSV file
3. Set File origin to 65001: Unicode (UTF-8)
4. Set Delimiter to Comma
5. Select LoadFor older versions of Excel, the Azure guide describes a Data tab option called From Text, which opens the Text Import Wizard. In the wizard you select Delimited, choose 65001 : Unicode (UTF-8) in the File origin field, select Next, choose the Comma option and select Finish.
65001 is the choice for a file saved as UTF-8. Microsoft's Text Import Wizard page gives the rule more generally: set File origin to the character set the file was made with, and its example is setting it to 1252 (Western European, Windows) when a file was made with that character set. If one choice still shows wrong symbols, try the other and compare the preview.
What Not to Do
Microsoft's Partner Center guide gives one explicit warning for files with non-English characters: do not convert the .csv file to the Excel file format (.xlsx) first, because the conversion can cause data loss or garbled output. Import the CSV with the correct File origin instead of opening it, saving it and hoping the symbols recover.
One limit worth knowing. Changing File origin can only decode characters that are actually in the file. If the symbols were already damaged when the CSV was written or re-saved by another program, no import setting can restore them. In that case go back to the original download or the original statement.
Check the Result Before You Rely on It
After loading, scan the description column for the accented names and the amount column for currency symbols. Look at a few rows against the source statement. If the data starts life in a PDF, PDFtoRow converts bank statement PDFs to Excel or CSV and shows a free preview of the first 10 rows, so you can check the characters before downloading.
Convert Bank Statement PDFs Cleanly with PDFtoRow
Tired of fixing broken columns, corrupted dates, and numbers stored as text? PDFtoRow extracts transaction tables directly from digital PDF vector layers, preserving check numbers and outputting true numeric amounts. Statements up to 5 pages are free once per email every 24 hours. Check the 10-row preview before you download.
Frequently Asked Questions
Microsoft describes File origin in the Text Import Wizard as the character set used in the text file. It is usually fine at the default, but you should set it to match the file if the file was created with a different character set from the one your computer uses.
Microsoft's Azure Cost Management and Partner Center guides both use 65001, shown as 65001: Unicode (UTF-8) in the Text/CSV dialog, for CSV files with accented or non-English characters.
No. Microsoft warns that converting the .csv to .xlsx can cause data loss or garbled output. Import the CSV with the correct File origin instead.
Primary Evidence & Official References
- Text Import Wizard — support.microsoft.com (Verified 2026-10-29). Official source documentation →
- Import or export text (.txt or .csv) files — support.microsoft.com (Verified 2026-10-29). Official source documentation →
- Downloads hub overview - Open a report in Excel that includes non-English character sets — learn.microsoft.com (Verified 2026-10-29). Official source documentation →
- Tutorial - Create and manage Cost Management exports (garbled characters FAQ) — learn.microsoft.com (Verified 2026-10-29). Official source documentation →