Collapse CSV rows into one cell per group
Paste a CSV, pick the columns to group by and the column to collapse, and get one row per group with that column's values joined into a single cell — the spreadsheet equivalent of SQL GROUP_CONCAT. Dedupe, sort, and choose the separator.
About this tool
Collapse rows groups a CSV by one or more key columns and joins another
column's values from every row in the group into a single delimited cell. It is
the spreadsheet-and-file equivalent of SQL GROUP_CONCAT / STRING_AGG /
LISTAGG, pandas groupby(keys)[col].agg(", ".join), and R dplyr
summarise(paste(x, collapse = ", ")).
Point it at the columns to group by and the column to collapse, and it returns
one row per group — the key columns followed by the collapsed column — in
first-seen order. You can dedupe repeated values (like DISTINCT), sort them
ascending or descending, skip blank cells, choose the separator between values,
and switch the field delimiter to tab, semicolon, or pipe. Everything runs
locally in your browser; the CSV never leaves your machine.
Worked example
Input:
region,product
East,Apple
West,Banana
East,Cherry
East,Apple
With group-by region, collapse product, and dedupe on:
region,product
East,"Apple, Cherry"
West,Banana
The two East rows merge into one; the repeated Apple is dropped by dedupe,
and the cell is quoted because it contains the comma separator.
Limits & edge cases
- Columns can be referenced by header name or 1-based index (
1= first column). Turn off First row is a header to treat every row as data — then use indices only, and output columns are labelledcol_1,col_2, …. - Output keeps only the key columns and the collapse column; any other columns are dropped.
- Groups and values within a cell are emitted in first-seen order unless you
choose a sort. Sorting is lexicographic (text), so
10sorts before2. - Skip blank cells (on by default) drops empty values before joining; turn it
off to keep empty positions (e.g.
a,,b). - The value separator can be any string (default
,). The field delimiter is separate and applies to both input parsing and output.
FAQ
How is this different from a group-by that sums or counts?
This tool concatenates the text values of one column into a list; it does not compute numeric aggregates like sum, average, or count. If you need those, use a CSV group-by / aggregate tool instead. Collapse rows is for turning many rows of labels, tags, or IDs into one comma-separated cell per group.
Can I group by more than one column?
Yes. Enter a comma-separated list in Group-by column(s) — for example
year,team. Rows are grouped by the exact combination of those key values, and
both key columns are carried through to the output.
Why is a collapsed cell wrapped in double quotes?
Standard CSV quoting. When the joined value contains the field delimiter (a comma by default), a line break, or a quote, the cell is wrapped in double quotes so the result re-parses correctly. Choosing a different separator or field delimiter can avoid the quoting if you prefer.
Does the order of the collapsed values match my input?
By default, yes — values appear in the order they were first seen, matching SQL
GROUP_CONCAT without an ORDER BY. Choose Ascending or Descending under
Sort collapsed values to order them alphabetically instead. Dedupe also keeps
the first occurrence of each value.
My file has no header row — can I still use it?
Yes. Turn off First row is a header. Then every row is treated as data, you
reference columns by 1-based index (1, 2, …), and the output gets generated
column labels col_1, col_2, … so the result is still a valid CSV.
Developer & Automation Access
Run it from the terminal
Same engine as this page, headless — via the gizza CLI:
gizza tool csv-collapse-rows "region,product
East,Apple
West,Banana
East,Cherry
East,Apple" 'key_columns=region' 'collapse_column=product'New to the CLI? Get gizza →
Open it by URL
Pre-fill and auto-run this tool with query parameters — the names match the API/CLI:
https://gizza.ai/tools/csv-collapse-rows/?data=region%2Cproduct%0AEast%2CApple%0AWest%2CBanana%0AEast%2CCherry%0AEast%2CApple&key_columns=region&collapse_column=product&separator=%2C%20&dedupe=true&skip_empty=true&sort_values=none&delimiter=comma&has_header=trueMachine-readable descriptor: tool.json — title + parameters JSON Schema for agents.