Double-clicking a CSV file is the fastest way to open it in Excel and the most common way to damage it. ZIP code 02134 turns into 2134. A 13-digit barcode becomes 4.00638E+12. Café shows up as Café. A file from a European system lands in one column because it uses semicolons. And if you save over the CSV afterwards, the damage is permanent.
This converter builds a proper Excel workbook from your CSV with each value stored the way you meant it. It reads the file in your browser, detects the delimiter and the encoding, keeps codes as text and real numbers as numbers, and writes an .xlsx file that opens the same way every time. Nothing is uploaded.
How to convert CSV to Excel
- Drop one or more CSV, TSV or TXT files on the box, or switch to Paste data and paste rows copied from anywhere.
- Check the file line: rows, columns, the detected delimiter (comma, semicolon, tab or pipe) and encoding. Change either one with the menus if it looks wrong.
- Look at the preview. Blue values will be stored as numbers and gold values will stay text. The stat boxes count both.
- Adjust the options: header row, leading zeros, numbers, thousands separators and dates.
- Click Download Excel (.xlsx). With several files, choose one workbook with a sheet per file, or separate .xlsx files in a ZIP.
The page loads with a short sample so you can see the result before using your own file. A message above the buttons tells you when you are looking at the sample.
What happens to each kind of value
The sample is a small product list. Here is how its values are stored with the default settings:
| Column | Value in the CSV | Stored in Excel as | Why |
|---|---|---|---|
| zip_code | 02134 | Text 02134 | Leading zero kept |
| barcode | 0012345678905 | Text 0012345678905 | Leading zeros and 13 digits kept |
| barcode | 4006381333931 | Text 4006381333931 | 12 or more digits stay text, so Excel cannot show it as 4.00638E+12 |
| price | "1,089.00" | Number 1089 | Thousands separator read |
| price | 42.5 | Number 42.5 | Plain number |
| stock | 0 | Number 0 | A single zero is a number |
| product | "Linen Throw, Oat" | Text Linen Throw, Oat | Quoted comma stays in the cell |
| launched | 2026-03-02 | Text, or a real Excel date with the date option | Dates are only converted when you ask |
Text cells that hold digits get Excel's Text number format (@), so if you edit them later Excel still keeps them as typed. Values that start with + are kept as text too, so phone numbers like +15035550142 keep their plus sign. Anything that starts with = is stored as text, never as a formula, which protects you from formula injection in files from unknown sources.
Which columns should stay text
A simple test: if you would never add the values up, they are labels, not numbers, and they belong in text cells. That covers more columns than people expect:
- ZIP and postal codes. US ZIP codes in New England and New Jersey start with 0, and so do many postal codes in France, Italy and Spain.
- Barcodes and GTINs. UPC (12 digits), EAN (13) and GTIN-14 codes often start with 0, and all of them are longer than Excel shows in General format. Marketplaces and Google Merchant Center reject codes that lost a digit; the GTIN lookup checks a code's check digit if you are unsure.
- SKUs, order numbers and account numbers. 00123 and 123 can be different products.
- Phone numbers. A leading + or 0 is part of the number.
- Credit card and bank references. These are 16 digits or more, and Excel would replace the last digits with zeros.
Quantities, prices, costs, weights and stock levels are real numbers, and the converter stores them as numbers so formulas and pivot tables work.
The options explained
| Option | Default | What it does |
|---|---|---|
| First row is a header | On | Makes row 1 bold, freezes it so it stays visible while you scroll, and adds filter buttons. |
| Keep leading zeros and long numbers as text | On | Values like 007, 02134 and +44 7700, and any value of 12 or more digits, stay text. Turn it off only if those columns really are quantities. |
| Store numbers as numbers | On | Lets you sum, sort and chart. Off stores every value as text, exactly as in the CSV. |
| Read 1,234.50 as a number | On | Accepts US-style thousands separators. |
| Decimal comma | Off | For European and Latin American files: reads 12,50 as 12.5 and 1.234,50 as 1234.5. Without it, decimal comma values stay text so nothing is misread. |
| Turn 2026-10-15 dates into Excel dates | Off | Converts ISO dates (and ISO date-times) into real Excel dates you can filter by month. Other date styles stay text, because 03/04/2026 means March in the US and April elsewhere. |
| Delimiter and Encoding | Detect | Override when detection guesses wrong, for example a single-column file. |
Column widths are set from the longest value in each column, up to a sensible maximum, so the workbook is readable as soon as it opens.
Encoding: why accents break and how this fixes it
A CSV file is plain text with no label saying which character encoding it uses. When you double-click one, Excel on Windows often assumes your system's legacy code page instead of UTF-8, so every accented letter, currency symbol and emoji comes out garbled. This converter checks the bytes first:
- A UTF-8 or UTF-16 byte order mark is read and removed.
- Files with the zero-byte pattern of UTF-16 are decoded as UTF-16. Some tools save "CSV" downloads as UTF-16 text separated by tabs, and those are detected too.
- Valid UTF-8 is used as UTF-8.
- Anything else is read as Windows-1252, the encoding older Excel versions and Windows apps use for Western languages.
If the preview still shows � symbols, choose the encoding by hand. Windows-1250 (Central European), Windows-1251 (Cyrillic), Shift_JIS (Japanese) and GBK (Simplified Chinese) are in the list. The workbook itself always stores text in Unicode, so the result looks the same on every computer.
Need a CSV, not a workbook?
Some systems only accept CSV. Download CSV for Excel rewrites your file as UTF-8 with a byte order mark and keeps your delimiter, which is enough for Excel to show accents correctly when the file is double-clicked. It cannot stop Excel from dropping leading zeros on open; only a real .xlsx can do that.
Example: a supplier price list from Europe
A French supplier sends tarifs.csv. Double-clicked on a US laptop it opens in one column, with Crème brûlée shown as Crème brûlée. Dropped here, the file line reads "3 rows, 4 col, semicolon, Windows-1252 (ANSI)": the converter found the semicolons and the old Windows encoding on its own. The preview shows the postal code 01000 in gold, so it stays text, and the price 12,50 as text too. Tick Decimal comma and 12,50 turns blue: it will be stored as the number 12.5, ready for a price comparison with your other suppliers. Download, and the workbook opens with the accents, the zeros and the prices right.
Why a real .xlsx beats a CSV for sharing
A CSV is only text, so every program that opens it has to guess again: the delimiter, the encoding, which columns are numbers. The guess depends on the computer's language settings, which is why the same file looks fine for you and broken for a colleague in another country. An .xlsx file stores those decisions inside the file. Text cells stay text, numbers stay numbers, the header stays frozen, and the file looks the same in Excel on Windows and Mac, Google Sheets and LibreOffice. Send the workbook to people who will read the data, and keep the CSV for systems that import it.
Several CSV files at once
Drop up to 50 files. In the default mode they become one workbook with one sheet per file, named after the file. Excel limits sheet names to 31 characters and forbids : \ / ? * [ ], so long or unusual names are shortened and cleaned, and duplicates get a number. Choose Separate .xlsx files (ZIP) if each file needs to stay on its own, for example one export per supplier. Click any file in the list to preview it.
A typical use: monthly keyword or sales exports from different tools, each with its own delimiter and encoding, combined into one workbook where every sheet has the same frozen header and filters. That beats importing each file through Excel's text import wizard one by one.
Limits
- Excel holds at most 1,048,576 rows and 16,384 columns per sheet. A larger file is refused with a clear message instead of being cut silently. Split it first.
- A cell can hold up to 32,767 characters; longer values are cut to that length.
- Quoted values that contain line breaks, such as product descriptions with several paragraphs, stay in one cell with the line breaks kept. Empty rows at the very end of the file are dropped.
- Files can be up to 100 MB each. Very large files depend on your device's memory, and 50 MB or more can take several seconds.
- CSV has no formatting, formulas or multiple tabs, so the workbook has none either, apart from the header styling, column widths and number formats described above.
- Dates other than ISO format stay text. Convert them in Excel with Data > Text to Columns if you need real dates.
Tip: To open a CSV in Excel without this tool and without damage, use Data > From Text/CSV instead of double-clicking. Set File Origin to 65001: Unicode (UTF-8), pick the delimiter, and under Data Type Detection choose "Do not detect data types" so codes keep their zeros. Recent Microsoft 365 versions of Excel also have File > Options > Data > Automatic Data Conversion, where you can stop Excel from removing leading zeros and converting long numbers.
Quick checks after converting
- Click a cell with a code such as a ZIP or SKU. The formula bar should show the leading zeros, and the cell should be left-aligned like text.
- Select a price column. The status bar should show a Sum; if it only shows a Count, those values were stored as text, usually because of a currency sign or a decimal comma.
- Scroll to the last row and compare the row number with the row count the converter showed.
- Search for the � character. If it appears, convert again with the encoding chosen by hand.
Common files people convert
Product exports from Shopify and WooCommerce, where SKUs and barcodes must survive; supplier price lists; keyword research exports, where search terms with accents and symbols need the right encoding; email lists from the email validator or the email extractor; and bank or payment exports. If you are planning content from keyword exports, Niche Research in AM Jarvis builds keyword maps from real Google searches without the spreadsheet step, and our guide to keyword research for niche sites explains how to group what you find.
Going the other way? The Excel to CSV converter exports any sheet of a workbook, or all of them as a ZIP, as clean UTF-8 CSV.