What is CSV?
Comma-Separated Values, a plain-text tabular format where each line is a row and columns are separated by a delimiter (usually a comma).
Also known as: Comma-Separated Values
CSV is the lowest-common-denominator data format. A header row names the columns, every subsequent row is a record, fields are separated by commas, and any field containing a comma, quote, or newline is wrapped in double quotes (with internal quotes doubled). Excel exports it, every database imports it, every language has a parser in the standard library.
The format is older than its standard (RFC 4180) and the standard is widely ignored. Real-world CSV uses tabs, semicolons, or pipes as delimiters; uses Windows or Unix line endings; sometimes has a BOM at the start; sometimes wraps every field in quotes; sometimes has trailing whitespace. A robust CSV parser sniffs the dialect rather than assuming.
For data exchange, CSV is still the default for tabular handoffs: bulk uploads to ad platforms, exports from CRMs, datasets shared between teams. Where it falls down is hierarchical data (nested objects), large strings (no compression, no streaming-friendly schema), and anything that needs strong typing. JSON Lines or Parquet are the typical replacements.
In the wild
- A nightly export of CRM contacts shipped as
contacts.csvwith 30,000 rows - An ad platform accepting a CSV upload of campaign updates and rejecting any row with malformed escaping
- A data team converting CSV to Parquet at ingestion time so analytics queries do not re-parse text