CSV to Excel Converter
Convert CSV files to Excel (XLSX) format.
Drop CSV file here
or click to select file
About CSV to Excel Converter
Drop a .csv file and get a real .xlsx workbook back. The file is read as UTF-8, the delimiter is detected from the content — comma, semicolon, tab or pipe — the rows are parsed into a worksheet named Sheet1, and the workbook is serialised and downloaded, all inside this browser tab with the data never sent anywhere. The picker only lists .csv files, so a tab- or pipe-separated export saved under another extension needs renaming first.
Why convert at all, rather than just opening the CSV
A CSV has no types, no formatting and no second sheet — it is text with separators, and every application that opens one has to guess what the columns mean. Converting to XLSX fixes the interpretation once, into a file that opens the same way for everyone you send it to. It also sidesteps the most irritating everyday failure: double-clicking a semicolon-separated export in Excel, which loads the whole thing into a single column because Excel expects the delimiter to match your system locale. Here the delimiter is detected from the file itself, so comma, semicolon, tab and pipe files all split correctly.
Read this before converting an ID column
Type inference is the part that quietly damages data, and it is worth being specific about how. Any field that looks numeric is stored as a number. A postcode of 01234 becomes the number 1234 — the original text is retained as the cell's display format so it still looks right, but sorting, filtering, lookups and any re-export all see 1234. Worse, numbers are held as doubles and are exact only to roughly 15 significant digits, so a 19-digit order reference or a long account number is silently rounded, with the final digits changed. If a column is an identifier rather than a quantity, that column needs to be text.
Dates carry the same class of risk in a different form. Date-looking text becomes an Excel date serial, and ambiguous slash-separated dates are interpreted month first — so 03/04/2024 lands as 4 March, not 3 April. Nothing warns you. Exports from systems outside the United States are the usual casualty. Writing dates in ISO form, year first, removes the ambiguity entirely and is worth doing at the source.
There is no import wizard here to override any of this, because the conversion is a single step rather than a dialogue. Two reliable workarounds: prefix a value with an apostrophe in the CSV to force it to text, or, when a whole file is full of identifiers, use Excel's own Text Import wizard where each column's type can be set explicitly before the data lands.
What the output workbook is and is not
It is a genuine XLSX — the zipped XML format Excel has used since 2007 — so it opens cleanly in Excel, LibreOffice Calc, Numbers and Google Sheets with no compatibility prompt. It contains one sheet, your rows in their original order, and nothing else: no bold header row, no frozen panes, no auto-fitted columns, no table styling, no formulas and no charts. Quoted fields are parsed properly, so a value containing commas, quotes or line breaks stays in one cell.
Encoding, size and privacy
The file is decoded as UTF-8 with any byte-order mark removed, which keeps accented names, CJK text and emoji intact; a file saved in a legacy single-byte encoding will arrive with replacement characters and should be re-saved first. Everything is held in memory at once and processed on the main thread, so low tens of megabytes are comfortable and much larger files will hang the tab. Because no request is made at any stage, a file full of customer records or payroll figures stays on your machine — which is the main reason to prefer a page like this over a service that asks you to upload first. To go the other way, use the Excel to CSV converter.
Finds The Delimiter Itself
Comma, semicolon, tab and pipe separated files are all detected automatically — useful because exports from European systems routinely use semicolons, which Excel then refuses to split into columns on import.
Read As UTF-8, Not Guesswork
The file is decoded as UTF-8 and any byte-order mark is stripped, so accented names, CJK text and emoji survive into the workbook instead of arriving as the mojibake a code-page guess produces.
Built In The Browser
The XLSX workbook is assembled in this tab and handed straight to your downloads. A customer list or a finance export is never transmitted, which matters more here than on most tools.
Frequently Asked Questions
My leading zeros are gone. What happened?
Anything that looks like a number is stored as one, so a postcode written 01234 is held as the number 1234 with 01234 kept only as its display format. It looks right on screen, but sort the column, run a formula against it or re-export and the zeros are gone. There is no import wizard here to mark a column as text. If leading zeros are load-bearing — postcodes, account numbers, product codes — prefix each value with an apostrophe in the CSV, or import the file through Excel's own Text Import wizard where you can set the column type.
A long ID column came out with the wrong digits.
That is floating-point precision, and it is worth knowing exactly where the cliff is. Numeric values are held as doubles, which are exact only to about 15 significant digits. A 19-digit identifier is therefore rounded — the last few digits change to zeros or shift by one — and the corruption is silent. Any identifier longer than 15 digits, which includes most credit-card numbers, many order references and all 64-bit IDs, must be treated as text rather than converted here.
How are dates handled?
Text that looks like a date is converted to an Excel date serial, and ambiguous formats are read as month-first. That means 03/04/2024 becomes 4 March rather than 3 April, which silently mangles data exported from anywhere using day-first ordering. Unambiguous ISO dates written year-month-day are safe. If your file uses day-first dates, convert them to ISO before running them through.
Which delimiters does it recognise?
Comma, semicolon, tab and pipe are all detected automatically from the content, so you do not need to tell it which you have. This matters most for semicolon files, which are the norm in locales where the comma is the decimal separator, and which Excel will happily open as a single unsplit column if you double-click them. The picker is narrower than the parser, though: it only accepts files ending in .csv, so a tab-separated .tsv or a plain .txt export has to be renamed before you can drop it in.
What does the resulting workbook look like?
One worksheet named Sheet1, containing your rows exactly as they were ordered in the file. There is no header styling, no frozen top row, no column auto-width, no table formatting and no formulas — it is your data placed into cells. The file is a genuine XLSX, so Excel, LibreOffice, Numbers and Google Sheets all open it without a compatibility warning.
Are quoted fields and embedded commas handled correctly?
Yes. Standard CSV quoting is parsed properly: a field wrapped in double quotes may contain the delimiter, line breaks and doubled quote characters, and all of it lands in a single cell. The common failure is a file that is not actually valid CSV — an unclosed quote somewhere in the middle will throw the parser out of alignment for every row after it.
Will accented characters and emoji survive?
Yes, provided the CSV really is UTF-8, which almost everything produces today. The file is decoded as UTF-8 and a byte-order mark is stripped if present. A file saved in a legacy single-byte encoding will come through with replacement characters, because there is no encoding selector — re-save it as UTF-8 first.
How big a file can it handle?
There is no coded limit and nothing is chunked: the whole file is read into memory as a string, parsed into a workbook object and serialised again, so peak memory runs to several times the file size. Files in the low tens of megabytes are fine on a desktop; well beyond that the tab becomes unresponsive, because the work happens on the main thread rather than a worker.
Can I convert several files at once?
No, one file per run — there is no multi-select and no way to merge several CSVs into one workbook with a sheet each. Convert them one at a time, or if you need many sheets in a single workbook, combine them in a spreadsheet application afterwards.
Do I need Excel installed?
Not to do the conversion, which happens entirely in your browser. You will need something that reads XLSX to open the result, but that includes free options — LibreOffice Calc, Google Sheets, Numbers on macOS, and Excel's own web version.