Merge versus join
Merge puts rows from multiple similar files into one dataset. Join combines columns from two datasets using an identifier such as customer ID, email, or SKU.
Stack versus union
Stack follows the first file’s column list and maps later rows by header name. Union keeps the superset of every header and leaves values blank where a file does not contain that column. Union is the safer choice when schemas differ.
Left versus inner join
A left join keeps every row from Dataset A, even without a match. An inner join keeps only rows whose key exists in both datasets. Swap the datasets when you need the other file to define the retained rows.
Why duplicate keys add rows
When one key appears several times in both datasets, preserving every match creates every valid pair. Two rows matched to three rows produce six output rows. This tool estimates the output first, reports duplicate keys, and blocks joins above its browser safety limit.
Private, bounded processing
Files remain in this browser tab. V1 accepts up to 5 MB per file, 20 MB combined, and renders only the first 50 preview rows. Downloaded output includes every processed row.