How to fix strange accent characters in a CSV

A CSV opens, but names and other accented text look wrong. You might see José instead of José, or a black diamond with a question mark where a letter should be. The file may still be intact, or its text may already have been changed.

First check a copy of the file, then import the original with the correct text encoding. The CSV Editor can help you inspect the rows and see which values are affected. It opens CSV files in your browser, and your files stay on your device.

Check the CSV text

1. Make a copy of the CSV so you can return to the original if needed.
2. Open the copy in the CSV Editor. Check the header and a few affected cells, including any names you can verify from another source.
3. Look closely at the broken text. Note whether you see extra characters such as Ã, or a replacement symbol such as �.
4. Open the original file through your spreadsheet app's import option, not by double-clicking it. In the import preview, choose UTF-8 if the app offers an encoding setting.
5. Check the preview before loading the data. If names look right, import the file and save a new copy as CSV UTF-8 if that format is available.

Importing is important because a spreadsheet may guess the encoding when you double-click a CSV. That guess can be wrong. Keep the original unchanged until you have checked the corrected copy.

What the strange characters mean

You see José or café

This usually means text saved as UTF-8 was read using a different encoding, often Windows-1252 or ISO-8859-1. UTF-8 is a common way to store text from many languages. When another encoding reads its bytes incorrectly, one accented character can turn into two or more visible characters. The original letters may still be recoverable by importing the original file with UTF-8.

You see a replacement symbol like �

A replacement symbol often appears when an app could not decode some bytes and substituted a placeholder. If the CSV was then saved with that placeholder, the original character may have been discarded. Try importing an untouched original or restoring a backup. Changing the encoding of the damaged copy usually cannot recreate the missing character.

You see even more extra characters, such as é

This can happen when text has been interpreted incorrectly and then saved again, a problem sometimes called double encoding. Go back to the earliest available copy and import that one. Each time a damaged version is opened and saved, the text can become harder to recover.

The text looks right in the preview

The data is probably intact, and the problem was how the app opened it. Import the file with the correct encoding, then export a new CSV using UTF-8. Reopen that new file to check that the accents still display properly.

How to fix the encoding

1. Import the original file. In Excel on Windows 11, select Data › From Text/CSV, choose the CSV, then check the preview. Set File Origin to 65001: Unicode (UTF-8) if the file is UTF-8 and the option is available. Do not load the file until the preview shows the accents correctly.
2. Use the import option on a Mac. In Excel for Mac, use Data › Get Data (Power Query) › From Text/CSV if those options are available, then select UTF-8 in the import settings. In older versions, look under File › Import for a CSV import option with an encoding choice. Menu names vary by version.
3. Try the source encoding if UTF-8 looks wrong. If the file came from an older Windows program, it may use Windows-1252. Select that encoding in the import preview if available. Use information from the person or system that created the CSV when you can, rather than trying random options on the only copy.
4. Check the imported values before saving. Compare several accented names with a reliable source. If the CSV contains codes such as 00127, set that column to text during import if the option is available. Spreadsheet apps can otherwise remove leading zeros or reinterpret values even when the encoding is correct.
5. Export a separate UTF-8 copy. Use the app's Save As or export option and choose CSV UTF-8 if it is listed. Keep the original file as a backup, then open the exported copy again to confirm the accents survived.
6. Ask for a fresh export if the characters are already lost. If every available copy contains � or the wrong letters, request a new CSV from the source system. Re-encoding a file cannot restore characters that were replaced or overwritten.

Why a CSV can look different in different apps

A CSV is plain text arranged into rows and columns. It does not always state which encoding should be used to read its text, so different apps may make different guesses. One app may display the accents correctly while another shows garbled characters, even though both opened the same file.

Changing the file extension or saving garbled text as UTF-8 does not fix the original interpretation. First read the source bytes using the right encoding, check the resulting text, and only then save a new UTF-8 CSV.

Use the CSV Editor to check the saved copy after import, but keep an untouched original until you have confirmed the repaired text.