Nothing is uploaded. When you drop or pick files, the tool reads each one locally with the browser’s File API, parses it in memory and computes the merge on your device — the row and column data never touch a network. There is no server, API, database or tracking in the processing path, so customer lists, financial exports or any other sensitive spreadsheet stay on your machine; the tool works offline once cached, and closing the tab discards everything. Even the Excel (.xlsx) export is built in the browser with a bundled library. The only optional network use anywhere on the site is consent-gated advertising, which never sees your data.

Append (a vertical merge) concatenates the rows of every file into one longer table — use it when the files describe the same kind of data, such as twelve monthly exports, and choose whether the output columns are the union of all files, only the first file’s headers, or must be identical. Side-by-side (a horizontal merge) puts each file’s columns next to the others and lines rows up by their position, padding shorter files with blanks. Join by column is the relational operation: you pick a key that appears in every file, and the tool matches rows sharing that key — an inner join keeps only matched keys, a left or right join keeps all rows from one side, and an outer join keeps everything, filling gaps with empty cells. If two files have a column of the same name, the join and side-by-side modes automatically rename the collision so no data is lost.

The parser inspects every file independently: it detects the delimiter (comma, semicolon, tab, pipe or caret), the character encoding (UTF-8, UTF-16 with a byte-order mark, or a Windows-1252 fallback) and whether a BOM is present, then reads the first row as the header and the rest as data, correctly handling quoted fields that contain commas or line breaks per RFC 4180. For an append merge you control headers directly — union includes every column from every file (blanks where a file lacks one), first-only trusts the first file’s header, and require-identical warns and falls back to a union if the headers do not match. You can also add a source_file column so each row records which file it came from. On export you choose the output delimiter, LF or CRLF line endings, whether to write a UTF-8 BOM (helpful for Excel), and whether to quote every field.

De-duplication runs after the merge so it can see the combined table. Keep all leaves every row untouched; Dedupe exact rows removes any row whose entire contents duplicate an earlier row; and Dedupe on key column keeps only the first occurrence of each value in a column you select, which is ideal for collapsing repeated customer IDs or e-mail addresses gathered from several files. The summary bar reports how many duplicate rows were dropped and the final row and column totals, and the live preview (up to the first 500 rows) reflects the result immediately, so you can confirm the merge before exporting to CSV, TSV, Excel, JSON or NDJSON.