When is this useful?
It is good for turning 1,000 detail rows into ten store totals, or a support log into counts by agent. That gives you a compact first look before deeper analysis.
Free web tool
Group CSV by a key column and calculate count, sum, average, min, or max.
Need sales by store or ticket count by owner? Group one column, choose a value column, and reduce a long CSV to SUM, AVG, COUNT, MIN, or MAX.
It is good for turning 1,000 detail rows into ten store totals, or a support log into counts by agent. That gives you a compact first look before deeper analysis.
COUNT counts rows in the group and does not need the value cell to be numeric. SUM, AVG, MIN, and MAX only use values that parse as finite numbers. So COUNT and AVG can be based on different numbers of rows.
A -, N/A, or blank-looking value is not treated as a valid measurement. If a group has no usable numeric values, MIN and MAX are left blank. A blank result therefore means “no valid number to aggregate,” not a numeric zero.
AVG is also blank when a group has no usable numeric values. SUM remains zero when there is nothing numeric to add. That difference is intentional: blank means “no valid value to summarize,” while zero is a numeric result.
Tokyo, tokyo, and a value with trailing spaces are separate groups. If near-duplicate labels appear in the result, normalize the source values before relying on the summary.
If a group has 100 rows by COUNT but far fewer usable numeric values, the value column may contain blanks or text. The aggregate is most useful when it sends you back to the exact part of the CSV worth checking.
The result does not list which source rows were skipped. When a number looks wrong, filter the original group and look for blanks, N/A, unit-suffixed text, or other values that did not parse as finite numbers.