Group & Aggregate a CSV

group by any columns · count · sum · avg · min · max · distinct · list

Collapse a CSV into one row per group and aggregate the columns you care about — total revenue per country, orders per customer, average response time per endpoint. It's the GROUP BY you'd write in SQL or the subtotal you'd build with a pivot table, without either.

group by sort output

group these columns

aggregate these columns

Or drop a .csv file here, or

ready

How to group a CSV

Paste the CSV or drop a file — the column checkboxes appear automatically.
Tick the columns to group by. Tick more than one for a compound group, like country + month.
Tick each column you want to aggregate and pick how: sum, avg, count, min, max, first, last, list, or distinct.
Every output row includes row_count, so you always know how many source rows each group represents.

What each aggregation does

count — non-empty values in the group. Blank cells don't inflate the number.
sum / avg — over values that parse as numbers; 1,200 is read as 1200, and non-numeric cells are ignored rather than counted as zero.
min / max — numeric when the column is numeric, otherwise alphabetical.
first / last — the value from the first or last row in the group, in file order. Handy for carrying along a name or ID that's constant within the group.
list — every value joined with | , so you can see what got rolled up.
distinct — how many different values the group contains, e.g. distinct customers per country.

Group by month, not by timestamp

Grouping on a raw timestamp gives one group per row, which is never what you want. Trim the value to the granularity you need first — Split column can cut 2026-03-14T09:12:00Z down to 2026-03 or 2026-03-14, and then grouping behaves. For anything more involved, SQL on a CSV lets you write the expression directly.

FAQ

How is this different from a pivot table?

Group-by produces one row per group with columns for each aggregate — a long, tidy table. A pivot table spreads one column's values across the top to make a matrix. Same data, different shape.

Can I sum a currency column with $ signs?

Strip the symbol first with Find & replace. Values like $1,200 aren't numbers; 1,200 is handled, $1,200 is not.

My groups are duplicated — "Kyiv" and "kyiv" separately.

Grouping is exact and case-sensitive. Normalise the column first — Trim whitespace for stray spaces, or Find & replace for casing.

Is anything uploaded?

No. Grouping happens in your browser; the file never leaves your machine.

Privacy

100% client-side. No upload. See the privacy policy.