JSON to CSV

Nested objects usually need to be flattened into column names such as address.city or address_postal_code. There is no single universal flattening rule, so check how the converter names nested fields before using the CSV in an import or reporting system.

Arrays are difficult to represent in a flat CSV row. A converter may join array values, serialize the array as JSON text, create multiple rows, or require you to flatten the data first. Choose the representation that matches the destination system.

The converter normally creates the union of discovered columns and leaves the missing cell empty for records that do not contain that field. This is common when API objects have optional properties.

For an array of similar objects, object keys generally become the header row. Nested keys may be flattened or transformed depending on the converter's rules.

CSV does not have a native type system, so values are ultimately written as text fields. Spreadsheet software may infer numbers, dates, or booleans when opening the file, and null is often represented as an empty field or a configured literal.

CSV fields containing commas, quotes, or line breaks should be quoted. Inside a quoted CSV field, double quotes are typically escaped by doubling them, following common CSV conventions documented in RFC 4180.

Yes if the CSV is exported using a Unicode encoding such as UTF 8 and the receiving application opens it with the correct encoding. Problems usually occur when spreadsheet software guesses a different character set.

Browser based conversion may become slow or run out of memory for very large JSON because the whole structure can be parsed and flattened in memory. For large production datasets, a streaming script or ETL process is safer.

CSV stores text fields, but Excel may automatically interpret them as dates, scientific notation, or numbers. If exact text must be preserved, import the CSV with explicit column types rather than double clicking it.

Not always. Flattening nested objects, arrays, null values, and data types can lose structural information. CSV is inherently tabular, so a round trip back to JSON may require custom reconstruction rules.