How to use the Text to Columns
Last verified 2026-08-09 · about 6 minutes to read
The short version
- 1.Paste your rows. The delimiter is detected automatically by looking for the one that produces a consistent column count.
- 2.Confirm the column count. The preview table shows how the data actually parsed. If a row has the wrong number of fields, the delimiter guess was wrong — set it manually.
- 3.Pick an output format. Markdown for documentation, JSON for code, SQL to seed a table.
Why this is worth getting right
Nearly every "text to columns" tool online is a call to split() on the delimiter. That is correct until a field contains the delimiter, which in real exports is immediately — every address, every "Surname, Firstname". This is a real RFC 4180 parser: quoted fields, doubled quotes, and newlines inside quotes all handled.
Method 1 — use the Text to Columns on this site
The fastest route. It runs entirely in your browser, so nothing is uploaded and there is no queue to wait in. No account is needed and there is no daily limit.
- Paste your rows. The delimiter is detected automatically by looking for the one that produces a consistent column count.
- Confirm the column count. The preview table shows how the data actually parsed. If a row has the wrong number of fields, the delimiter guess was wrong — set it manually.
- Pick an output format. Markdown for documentation, JSON for code, SQL to seed a table.
Method 2 — do it without this site
Worth knowing, because a tool you cannot replace is a dependency rather than a convenience. Most tasks in the text division have a command-line or built-in equivalent; it is usually more setup and less convenient, but it works offline and it is scriptable, which matters once you are doing something a hundred times instead of once.
What this tool will not do
Every tool has an edge. These are ours for text to columns, stated up front so you find out here rather than halfway through a deadline:
- Space-delimited input cannot distinguish a separator from a space inside an unquoted field. Use tabs where you can.
- SQL output infers only "looks like a number" versus "string". It does not detect dates, booleans or nulls beyond empty fields.
- Fixed-width column input is not supported — this splits on a delimiter, not on column positions.
Questions people ask
Why does splitting on a comma break my data?
Because a field can contain a comma. `"Rahman, Ayesha",Engineer` is two fields, not three — the quotes say so. A plain split produces three, silently shifting every column after it. This tool implements RFC 4180 quoting, so quoted fields survive intact.
How does automatic delimiter detection work?
By consistency rather than frequency. Counting occurrences picks whichever character is most common, which on prose is the space. Instead it tries each candidate and keeps the one that produces the same number of fields on every line — which is what a delimiter actually does.
How do I put a quote inside a quoted field?
Double it. `"She said ""hello"""` is one field containing `She said "hello"`. That is the RFC 4180 rule, and this parser follows it, as does Excel.
Can fields contain line breaks?
Yes, if they are inside quotes. The parser reads the whole document rather than splitting on newlines first, so a quoted field spanning two lines is kept as one value.
Are the SQL statements safe to run?
They are escaped — single quotes are doubled and numbers are left unquoted — but treat them as a starting point, not as trusted input. Read them before running them against anything that matters, and never build a production import from a browser tool without checking it.
What happens if rows have different column counts?
They are kept as they are and the mismatch is reported above the result. Ragged rows are usually a sign the delimiter guess was wrong, or that a quote was left unclosed somewhere in the file.