CSV to JSON for Every Role: 6 Real Uses, Made Easy

CSV to JSON sounds like one job, yet a tester, an analyst and a shop owner need three different results from the same file. The right output shape and one checkbox decide whether your data arrives clean or quietly damaged. Good csv to json work is mostly about those two choices.

Below you will find one sample file, the exact JSON it produces under each setting, and six role based playbooks for csv to json work. Skip to your role if you are in a hurry.

What CSV to JSON conversion changes

CSV is a grid of text. Every cell is a string, and the file has no idea whether 42 is a number or a label. JSON adds structure and types: objects, arrays, numbers, booleans and null. Conversion is the act of guessing those types and choosing a shape, and every csv to json tool does both.

Here is a small file with a trap in it, a perfect test for csv to json settings. The first customer has a comma inside the name, a leading zero in the ID, and the second customer has an empty balance.

Sample CSV for the csv to json test

Copyid,name,city,balance,active
001,"Smith, John",Berlin,1200.50,true
002,Ana Lopez,Madrid,,false

Array of objects, type inference on, empty cells as null

Copy[
  {"id": 1, "name": "Smith, John", "city": "Berlin", "balance": 1200.5, "active": true},
  {"id": 2, "name": "Ana Lopez", "city": "Madrid", "balance": null, "active": false}
]

Array of objects, type inference off

Copy[
  {"id": "001", "name": "Smith, John", "city": "Berlin", "balance": "1200.50", "active": "true"},
  {"id": "002", "name": "Ana Lopez", "city": "Madrid", "balance": "", "active": "false"}
]

Notice what moved. With inference on, 001 became 1 and 1200.50 became 1200.5. With it off, every value stays text, which is safe but means your code must convert numbers later. Neither csv to json result is wrong. Each fits a different job.

Six real jobs, by role

Pick the tab that looks like your day, then see how csv to json fits it. Each one names the task, the setting to choose and the mistake to avoid.

Choosing the output shape for your job

Most people leave the shape at its default and never look back. That works until the data is large or the consumer is picky. Use this table to match the shape to the receiver.

File size is a real consideration. Array of objects repeats every key name in every row. A file with 10,000 rows and 12 columns repeats each header 10,000 times. Array of arrays stores the header once, so it is noticeably smaller, at the cost of readability.

Type inference: when to turn it on or off

Type inference tests each trimmed value in a fixed order. Exact true or false becomes a boolean. Exact null becomes JSON null. Anything that passes a number test becomes a number. All else stays a string.

Delimiters, quotes and line endings

The name says comma, but real files use whatever the exporting program chose. Spreadsheets in many European locales write semicolons because the comma is already the decimal mark. Database dumps often use tabs or pipes.

Quotes follow RFC 4180, the closest thing CSV has to a standard. A field with a comma, a line break or a double quote should be wrapped in double quotes. A quote inside such a field is written twice.

Quoting rules in one line each

Copyplain text          -> plain text
Smith, John         -> "Smith, John"
5" pipe             -> "5"" pipe"
line one + line two -> "line one
line two"

A checklist before you convert

The tool shows rows, columns, key count and output size after each run. Compare the row count with your spreadsheet. A mismatch usually means a stray quote swallowed several lines into one cell.

Convert a file step by step with csv to json

1. Load the csv to json dataPaste the text, pick a sample, or choose a file with Upload. The file is read in your browser and goes nowhere.
2. Set the delimiter and headerChoose comma, semicolon, tab, pipe or a custom character. Keep first row is header on unless the file has none.
3. Choose the output modeArray of objects, array of arrays, or keyed by first column. Use the table above to decide.
4. Set the cleaning optionsTick trim whitespace, skip empty rows and empty as null where they fit. Decide on inference using the label test.
5. Convert and copyPress convert, check the counts, then copy the result or download the file. The highlighted view makes errors easy to spot.

Need the reverse trip later? The JSON to CSV converter turns an array of objects back into rows for a spreadsheet.

Questions people ask about CSV to JSON

Sources: RFC 4180, Common Format and MIME Type for CSV Files