What is the difference between stacking and joining CSV files?

Stacking appends every row from each file into one taller table, lining columns up by header name. Joining matches rows across files on a shared key column (such as an id) and widens the table by adding the other file's columns to matching rows.

What do inner, left and full join mean?

An inner join keeps only rows whose key exists in both files. A left join keeps every row from file A and pulls in matching columns from file B where the key matches. A full join keeps every row from both files, matched on the key where possible and left blank where not.

Can I merge more than two CSV files at once?

Yes. Stacking unions any number of files. Joins fold left to right — A joins B, then that result joins C, and so on — reusing the same key column name where it exists in each subsequent file.

What happens if two files have a column with the same name?

When stacking, identically named columns are merged into one. When joining, a colliding column from the second file is kept and renamed with a suffix (for example status_b2) so no data is silently overwritten.

Are my CSV files uploaded to a server?

No. Parsing, merging and export all happen locally in your browser using the PapaParse library. Your data never leaves your device, which makes the tool safe for confidential exports.

Why are some cells blank in my merged file?

Blank cells appear where a row had no value for a column that only exists in another file. In a left or full join, rows with no match on the other side keep their own columns and leave the joined columns empty.

CSV Merger — Gera Tools

CSV Merger combines two or more CSV files into a single table — either by stacking all the rows on top of each other or by joining them on a shared key column. It is for anyone who has split data across several exports and needs one clean file: a marketer reconciling sign-ups with plan data, an analyst combining monthly reports, or a developer stitching together API dumps without opening a spreadsheet. Upload your files or paste the text, choose how to merge, preview the result and export. Nothing is uploaded — the whole process runs in your browser, so even confidential customer lists stay on your machine.

How it works

Each file is parsed with PapaParse, treating the first line as the header row. You then choose a merge mode:

Stack appends every row from every file into one table. The output columns are the union of all headers, so a column that exists in only one file is kept and left blank for rows from the others. An optional pass removes exact duplicate rows.
Inner join keeps only rows whose key value appears in both files, combining their columns side by side.
Left join keeps every row from file A and adds file B’s columns wherever the key matches, leaving them blank otherwise.
Full join keeps every row from both files, lined up on the key where possible.

For joins you pick the key column in each file — they do not need the same name, so you can match email in one file against user_email in another. Optional toggles trim surrounding whitespace and make the key match case-insensitive. When a non-key column name collides between files, the second file’s copy is renamed with a suffix rather than overwritten, so no values are lost. With three or more files, joins fold left to right.

Example

Suppose people.csv has id,name,email and plans.csv has id,plan,signup. Choosing a left join on id keeps all three people and attaches each one’s plan:

id	name	email	plan	signup
1	Ava Stone	[email protected]	Pro	2026-01-04
2	Ben Cole	[email protected]
3	Cara Diaz	[email protected]	Free	2026-02-19

Ben has no row in plans.csv, so his plan columns are blank — but he is still kept because a left join preserves every row from file A. Switch to an inner join and Ben drops out; switch to a full join and the unmatched plan for id 4 is appended too. Press Download merged CSV to save the combined file, ready to open in Excel, Google Sheets or any database import. Every figure here is computed locally — no numbers are uploaded or stored.