CSV ↔ JSON Converter: Frequently Asked Questions

Empty cells (two consecutive delimiters with nothing between) become empty strings in JSON by default. With type inference enabled, they remain empty strings rather than being converted to null, because an empty field is ambiguous — it might mean null, zero, false, or a genuinely empty string depending on context.

Yes — uncheck 'First row is header'. The converter will treat all rows as data rows and produce a JSON array of arrays rather than an array of objects.

The converter handles arrays of objects (most common API format) and arrays of arrays. It does not handle deeply nested objects, objects with array values, or single objects (non-array JSON). For these structures, the output will be incomplete or incorrect.

Type inference only converts values that parse unambiguously as numbers. Values like '1,234' (with comma thousand separator) or '€30' (with currency symbol) remain strings because they are not valid JSON numbers. Remove formatting before converting, or disable type inference and parse manually in your application.

The converter runs in your browser using JavaScript. Very large files (over ~10MB) may be slow or cause browser memory issues. For large files, use a server-side tool like Python's pandas library (pd.read_csv() and .to_json()) or command-line tools like jq.

Not directly. Nested objects cannot be cleanly represented as flat CSV columns. Flatten the JSON first — convert nested keys to dot-notation strings (address.city, address.country) — before converting to CSV.