CSV to JSON Converter

Convert your CSV data to JSON format with our powerful online tool. Features include custom field mapping, nested JSON support, and real-time preview.

Loading editors...

Auto Sync

Automatically sync from left

Loading editors...

Want the background behind this tool? CSV Data Processing: Parsing, Converting, and Best Practices

You exported a spreadsheet, and the next thing in the pipeline — a React component, a Postgres seed script, or a Node CLI — wants JSON instead of rows. Hand-converting works for the first ten rows and falls apart after that. Paste the CSV here and the right pane shows the JSON array of objects, with the header row driving the property keys.

The parser implements RFC 4180 through PapaParse, which means quoted fields containing commas, embedded newlines, and escaped double quotes all round-trip correctly. Conversion runs entirely in your browser — useful when the CSV contains user records, order history, or anything else you would rather not ship to a third-party server.

How to use

Paste CSV into the left editor, or upload a .csv file.
The first row becomes the JSON property keys automatically.
PapaParse auto-detects the delimiter (comma, semicolon, tab, or pipe) from the first few rows.
Numeric and boolean values come through as their native JSON types, not quoted strings.
Copy the JSON or download it as a .json file.

Examples

The header row becomes object keys and every data row becomes one element of the resulting array.

Example 1: Header row becomes object keys

Input

name,age,city
Alice,30,Seoul
Bob,25,Tokyo
Carol,41,Berlin

Output

[
  { "name": "Alice", "age": 30, "city": "Seoul" },
  { "name": "Bob", "age": 25, "city": "Tokyo" },
  { "name": "Carol", "age": 41, "city": "Berlin" }
]

Example 2: Quoted field with embedded comma

When a value itself contains a comma, the CSV wraps it in double quotes. The parser keeps that value intact instead of splitting on the inner comma.

Input

id,full_name,note
1,"Smith, John","VIP, renewed"
2,"Doe, Jane","lapsed"

Output

[
  { "id": 1, "full_name": "Smith, John", "note": "VIP, renewed" },
  { "id": 2, "full_name": "Doe, Jane", "note": "lapsed" }
]

CSV specification reference

Quick reference from RFC 4180. The converter enforces these rules on input; anything outside them is rejected or normalized.

Element	Meaning	Example
Record	One line, terminated by CRLF	a,b,c
Field	Separated by a comma (or custom delimiter)	foo
Quoted field	Double-quotes wrap values with commas or CRLF	"a,b"
Escaped quote	Double-quote doubled inside a quoted field	"he said ""hi"""
Header	Optional first record listing field names	name,age

Common invalid forms

a,"b   // unterminated quoted field
a,b,"c,d  // missing closing quote on last field
a,	b,c  // mixed tabs are a dialect, not RFC 4180

RFC 4180 quoting rules (the parts that bite)

Wrap any field that contains a comma, double quote, or newline in double quotes. Bare commas inside a field cause every parser to split it into two columns.
Escape a literal double quote inside a quoted field by doubling it — "She said ""hi""" produces the JSON value She said "hi". A single backslash is not an escape in CSV.
Newlines inside quoted fields are allowed and translate to "\n" in the JSON string. The parser uses the quote state machine to keep them inside the cell.
Trailing whitespace inside quotes is preserved. Outside quotes it is usually preserved too, which is why exports from older systems often leave stray spaces around values. Use the CSV Formatter to strip them.

Excel and locale gotchas

European Excel uses semicolons instead of commas because the comma doubles as the decimal separator. Paste a German or French export and the parser will pick up the semicolon automatically.
Excel for Windows writes UTF-8 BOM (EF BB BF) at the start of the file when you choose "CSV UTF-8". PapaParse strips it silently; a naive Python open() on the same file will leave a literal three-byte prefix on the first header name.
Long numeric IDs lose precision when Excel re-opens the CSV — IDs like 1234567890123456 get cast to scientific notation and the last digits become zero. Convert directly from the source, not from the file you opened and saved in Excel.
Copy from a cell range produces TSV, not CSV. Tabs as separators with no surrounding quotes — the parser handles it but you can't paste the same text into a tool that strictly expects commas.

Frequently asked questions

Why are my numeric IDs coming through as strings?

Two reasons: either the value contains a leading zero (which the parser preserves as a string so you don't lose the zero), or the value exceeds Number.MAX_SAFE_INTEGER (2^53 − 1). JavaScript numbers cannot hold a 16-digit integer without precision loss, so the converter keeps them as strings. Cast them to BigInt on the consumer side if you need integer math.

Can I build nested JSON objects from a flat CSV?

Not out of the box. The output is always a flat array of flat objects keyed by header. To nest, name your columns with dot notation (address.city, address.zip) and run a post-processing step like Lodash _.set after conversion, or use a CSV with a JSON column and parse that column manually.

What if my CSV has no header row?

You have two choices. Add a synthetic header row before pasting (it's the cleanest fix and only takes one line). Or paste as-is and the converter will produce objects keyed by the first data row's values, which is rarely what you want — generic col_1, col_2 keys are better added by hand.

Are duplicate header names allowed?

CSV allows them; JSON does not. When two columns share a header, the second value overwrites the first in the resulting object — so a CSV with two email columns will only carry the second one through to JSON. Rename duplicates in the source before converting.

Can I paste data directly from Excel or Google Sheets?

Yes. Copying cells from a spreadsheet puts tab-separated values on the clipboard. The auto-detection picks up the tab delimiter and parses the rows correctly. If your spreadsheet contains formula references, copy as values first (Cmd/Ctrl + Shift + V in Sheets) so the JSON gets the computed numbers instead of =SUM(...) strings.

Related guides

JSON vs YAML: When to Use What — A Developer's Guide

JSON wins on APIs; YAML wins on configs. Side-by-side syntax, parser behaviour, and where each fits across Kubernetes manifests, REST payloads, and GitHub Actions.

CSV Data Processing: Parsing, Converting, and Best Practices

CSV looks trivial until a field contains a comma in quotes inside a newline. Format rules, parser pitfalls, language-by-language code, and a checklist for safe conversions.

JSON Schema Validation: Protecting Your API with Data Contracts

TypeScript types vanish at runtime, so your API still trusts whatever shows up in req.body. JSON Schema 2020-12, Ajv, and how to ship contracts that hold under real traffic.

JSON Diff Workflows for API Change Reviews

A practical workflow for comparing JSON payloads, spotting contract-breaking API changes, and deciding when a schema update is safer than a manual review.

CSV Data Processing: Parsing, Converting, and Best Practices10 min read

CSV to JSON Converter

Convert your CSV data to JSON format with our powerful online tool. Features include custom field mapping, nested JSON support, and real-time preview.

Loading editors...

Auto Sync

Automatically sync from left

Loading editors...

Want the background behind this tool? CSV Data Processing: Parsing, Converting, and Best Practices

How to use

Paste CSV into the left editor, or upload a .csv file.
The first row becomes the JSON property keys automatically.
PapaParse auto-detects the delimiter (comma, semicolon, tab, or pipe) from the first few rows.
Numeric and boolean values come through as their native JSON types, not quoted strings.
Copy the JSON or download it as a .json file.

Examples

The header row becomes object keys and every data row becomes one element of the resulting array.

Example 1: Header row becomes object keys

Input

name,age,city
Alice,30,Seoul
Bob,25,Tokyo
Carol,41,Berlin

Output

[
  { "name": "Alice", "age": 30, "city": "Seoul" },
  { "name": "Bob", "age": 25, "city": "Tokyo" },
  { "name": "Carol", "age": 41, "city": "Berlin" }
]

Example 2: Quoted field with embedded comma

When a value itself contains a comma, the CSV wraps it in double quotes. The parser keeps that value intact instead of splitting on the inner comma.

Input

id,full_name,note
1,"Smith, John","VIP, renewed"
2,"Doe, Jane","lapsed"

Output

[
  { "id": 1, "full_name": "Smith, John", "note": "VIP, renewed" },
  { "id": 2, "full_name": "Doe, Jane", "note": "lapsed" }
]

CSV specification reference

Quick reference from RFC 4180. The converter enforces these rules on input; anything outside them is rejected or normalized.

Element	Meaning	Example
Record	One line, terminated by CRLF	a,b,c
Field	Separated by a comma (or custom delimiter)	foo
Quoted field	Double-quotes wrap values with commas or CRLF	"a,b"
Escaped quote	Double-quote doubled inside a quoted field	"he said ""hi"""
Header	Optional first record listing field names	name,age

Common invalid forms

a,"b   // unterminated quoted field
a,b,"c,d  // missing closing quote on last field
a,	b,c  // mixed tabs are a dialect, not RFC 4180

RFC 4180 quoting rules (the parts that bite)

Wrap any field that contains a comma, double quote, or newline in double quotes. Bare commas inside a field cause every parser to split it into two columns.
Escape a literal double quote inside a quoted field by doubling it — "She said ""hi""" produces the JSON value She said "hi". A single backslash is not an escape in CSV.
Newlines inside quoted fields are allowed and translate to "\n" in the JSON string. The parser uses the quote state machine to keep them inside the cell.
Trailing whitespace inside quotes is preserved. Outside quotes it is usually preserved too, which is why exports from older systems often leave stray spaces around values. Use the CSV Formatter to strip them.

Excel and locale gotchas

European Excel uses semicolons instead of commas because the comma doubles as the decimal separator. Paste a German or French export and the parser will pick up the semicolon automatically.
Excel for Windows writes UTF-8 BOM (EF BB BF) at the start of the file when you choose "CSV UTF-8". PapaParse strips it silently; a naive Python open() on the same file will leave a literal three-byte prefix on the first header name.
Long numeric IDs lose precision when Excel re-opens the CSV — IDs like 1234567890123456 get cast to scientific notation and the last digits become zero. Convert directly from the source, not from the file you opened and saved in Excel.
Copy from a cell range produces TSV, not CSV. Tabs as separators with no surrounding quotes — the parser handles it but you can't paste the same text into a tool that strictly expects commas.