Convert CSV to JSON
Free online converter for CSV files. Structured data interchange format. Nothing to install, no watermark, and every file is deleted automatically.
Drag and drop your files
or
Upload a CSV file to convert it to JSON. Up to 3 files at once, 100 MB each.
Free · up to 3 files · 100 MB each · files deleted automatically after 1 hour.
About CSV and JSON
CSV .csv
Rows of plain text with values separated by commas. No formatting, no formulas, no sheets — just the data, which is what makes it universal.
Good for
- Readable by every spreadsheet, database and programming language
- Tiny, and trivial to generate or parse
- Nothing hidden — the file is exactly what it appears to be
Trade-offs
- One table only: no sheets, formulas, colours or charts
- No standard for encoding, so accents and symbols often corrupt
- Commas and line breaks inside values need careful quoting
text/csv
JSON .json
A structured text format for nested data. Unlike CSV it represents hierarchy, which is why nearly every web API speaks it.
Good for
- Represents nesting and lists, not just flat rows
- Parsed natively by essentially every language
- Readable, and precise about types
Trade-offs
- More verbose than CSV for plain tables
- Awkward to read in a spreadsheet without flattening
- No comments, and no native date type
application/json
CSV and JSON side by side
| CSV | JSON | |
|---|---|---|
| Compression | None — plain text | None — plain text |
| Animation | No | No |
| Typically used for | Data exchange, imports and exports, reporting | APIs, configuration, structured exports |
What changes when you convert CSV to JSON
Rows become records
The first row is read as field names, and every row after it becomes one object keyed by those names. That gives you the array of records most APIs and scripts expect. A sheet whose real header does not sit in the first row will produce field names taken from whatever does — worth checking before you point code at the output.
Types are inferred, not declared
JSON distinguishes the number 42 from the string "42"; a CSV cannot. Values that read as numbers are written as numbers, and everything else stays a string. Where that guess matters to you — identifiers with leading zeros, telephone numbers, anything you never want arithmetic performed on — check those fields in the result.
Read directly, never opened
No spreadsheet application is involved in this conversion. The file is parsed in our own process and the data is written straight out, which is both faster than launching an office suite and safer: macros in a workbook are never in a position to run, because nothing here is capable of running them. It also means the output is exactly the data, with none of an office suite's opinions about formatting applied on the way through.
Read as UTF-8
A CSV is a text file, and a text file does not record which encoding it uses — that is the one thing the format cannot tell you about itself. We read it as UTF-8, which is right for essentially anything produced this decade. A file saved long ago in a regional encoding can arrive with accented characters mangled, and the fix is to re-save it as UTF-8 before converting rather than after. Tabs and spaces used to line columns up will also shift once the text is set in a proportional typeface.