CSV Delimiter and Encoding Reference
By CSV Editor Team · Last updated: 2026-03-16
The short answer is: CSV files usually break because the parser expects the wrong delimiter or the text is opened with the wrong encoding. If you understand those two settings, you can prevent most “one long column,” “garbled characters,” and “import failed” issues before they spread downstream.
This guide explains when to use commas, semicolons, or tabs, when UTF-8 is the right choice, what BOM means, and how to make cleaner import decisions across spreadsheets, CRMs, ecommerce tools, and custom systems.
Delimiter basics: comma, semicolon, and tab
Comma-delimited CSV is the default most people mean when they say “CSV.” It is common in web apps, APIs, and many import templates.
Semicolon-delimited CSV is common in locales where commas are used as decimal separators. A spreadsheet may open semicolon-based files correctly while a web importer expects commas, or vice versa.
Tab-delimited files (TSV) are useful when the data contains many commas and you want a cleaner separator. They are still structured text, but technically not comma-separated values.
Encoding basics: UTF-8, BOM, and legacy file problems
UTF-8 is the safest modern default for multilingual text, symbols, and exports that move between systems.
UTF-8 with BOM adds a marker at the beginning of the file. Some older Windows-oriented tools and spreadsheet workflows expect it to detect UTF-8 correctly.
Legacy encodings such as Windows-1252 or ISO-8859-1 can still appear in exported files. If a file was saved in one encoding and opened as another, text may render as mojibake like é or replacement characters like �.
How delimiter and encoding problems show up in practice
| Symptom | Likely setting issue | What to do |
|---|---|---|
| CSV opens as one giant column | Wrong delimiter | Match the parser to comma, semicolon, or tab |
| Text appears as Ã, ’, or � | Wrong encoding | Re-open as UTF-8 or the expected legacy encoding |
| Descriptions split across columns | Quoted field parsing issue | Check quotes before changing delimiters blindly |
| Import succeeds but text is corrupted | Encoding mismatch on export | Export using the destination system’s expected encoding |
A safe workflow for choosing the right settings
The most reliable approach is to decide settings from the destination backwards: what does the importer expect, what does the source file actually contain, and which sample rows prove you chose correctly?
- Check what the destination importer explicitly requires.
- Open the source file with the delimiter that makes rows and columns render correctly.
- Verify a few rows with quoted commas, semicolons, tabs, or line breaks inside cells.
- Review non-English text, symbols, and punctuation to catch encoding problems early.
- Export a test file with the chosen delimiter and encoding.
- Run a sample import before processing the full dataset.
Example: EU spreadsheet export into a US web app
A common failure case is a semicolon-delimited CSV exported from a European spreadsheet and uploaded into a US web app that expects commas. The app then imports the whole row into one column and may also misread accented characters if encoding changes happened during export.
The fix is usually straightforward: parse the original file with semicolons, confirm UTF-8 text is intact, then re-export as comma-delimited UTF-8 if that is what the destination expects.
Quick tips
- Delimiter and encoding are separate settings, but they often fail together.
- Do not assume “CSV” always means comma-delimited.
- UTF-8 is the safest default unless the destination explicitly says otherwise.
- Keep a sample row with special characters for export QA.
FAQ
Is semicolon-delimited CSV still a CSV file?
Yes. Many tools still call it CSV even though the separator is a semicolon. The important thing is whether the destination parser expects that format.
Should I use UTF-8 with BOM?
Only when a destination tool or spreadsheet workflow specifically benefits from it. For many modern web apps, plain UTF-8 works well without BOM.
What is the fastest way to verify delimiter and encoding before import?
Open the file in a CSV-aware editor, check whether the table structure is correct, and verify a few rows containing accents, symbols, quotes, and commas inside text fields.
Related guides
Canonical: https://csveditoronline.com/docs/csv-delimiters-encoding