Merge CSV Files

updated 12 March 2026

Before you start

You'll need:

• Two or more CSV files, each with a header row. The tool reads headers to figure out columns.
• For Outer-join, a column name that exists in every file and uniquely identifies a record (typically id, user_id, email, etc.). Leading/trailing whitespace is trimmed on the key; casing is compared as-is.

Files don't have to have the same columns. Concatenate mode builds a column union; outer-join mode does the same for non-key columns.

How to use it

1. Drop all your CSV files onto the drop zone (multi-select from the picker also works). The file list on the left shows what's loaded.
2. Pick Concatenate rows or Outer-join on key.
3. For outer-join, type the exact column name in Join key. It's case-sensitive.
4. Click Merge. The combined CSV appears on the right; the status bar shows the row and column count.
5. Click Download .csv to save, or copy the output.
6. Use Clear files to reset.

Modes explained

Concatenate rows

Stacks all rows from every file on top of each other, preserving file order. The output has the union of all headers seen across the inputs; columns missing from a given file become empty cells in its rows.

When to use it: one export split across months, shards, or regions; daily dumps you're combining into an archive; CSVs from different teams with mostly-aligned schemas.

Outer-join on key column

Treats one column as the record identifier. Rows sharing the same key across files are merged into one output row, column-by-column. Rows whose key cell is empty are skipped (with a count in the status bar). If two files have a non-key column with the same name, later files overwrite earlier values for the same key — watch for this if your "same" column actually has different meanings in different exports.

When to use it: combining users + orders + subscriptions by user_id; enriching a product list with inventory / pricing / translations; merging survey responses across question batches by respondent_id.

Example

users.csv:

id,name
1,Alice
2,Bob

orders.csv:

id,total
1,42
3,17

Outer-join on id:

id,name,total
1,Alice,42
2,Bob,
3,,17

Tips & common pitfalls

• Join keys must match exactly. "42 " (with a trailing space) and "42" are two different keys unless you pre-trim. The tool trims leading/trailing whitespace on keys, but nothing fancier.
• Duplicate keys within a file produce one output row per combination; you may end up with more rows than you expect. Dedupe first using Dedupe CSV if that's a problem.
• Pick files in the right order for concatenate. The UI merges in the order files were added. Clear and re-add if you need a specific order.
• Header spelling matters. email, Email, and e_mail become three separate columns in concatenate mode. Normalise headers before merging if your exports are inconsistent.
• Huge outer-joins allocate a hash map keyed by the join column; expect the memory footprint to be a few multiples of the raw file size.

Troubleshooting

Frequently asked questions