Find added, removed and changed rows between two CSV or Excel files, cell by cell
Processed in your browser. Nothing is uploaded. Checked against sources on 17 Oct 2026.Compare Two CSV Files matches the rows of two CSV or Excel files (up to 50 MB and 500,000 rows each) on a key column such as an ID or email, then lists every added, removed and changed row, with each changed cell before and after. It runs in your browser and nothing is uploaded.
Press Select two files and pick the older and the newer version, CSV or .xlsx; they are read on your device.
Check the key column, such as id or email, that says which rows are the same record.
Read the counts of added, removed and changed rows, and each changed cell before and after.
Press Download to save the list of differences as a CSV for Excel.
A line-by-line diff says line 812 is different. That is little help when the export was sorted differently this month, or a column was added in the middle. This tool compares records, not lines.
How rows are matched. One column is the key: the value that says two rows are the same record. A column named like an ID (id, code, sku, email, order id) whose values are unique in both files is chosen for you; otherwise the first unique, filled column. You can pick another, or compare by row position when there is no key at all.
What you see.
| Result | Meaning |
|---|---|
| Added | The key is only in the second file |
| Removed | The key is only in the first file |
| Changed | The key is in both and at least one cell differs; each cell is shown old and new |
| Unchanged | Every compared cell is equal |
Columns are matched by their header name, so a column that moved is still compared. Columns in only one file are named and left out of the comparison, so a new column does not mark every row as changed.
What counts as equal. By default, spaces at the ends are ignored, and numbers are compared as numbers, so 1.50, 1.5 and 1,000 against 1000 are equal. Upper and lower case can be ignored too, which helps with email lists.
Odd rows are named. A key that appears twice in one file is listed, and only its first row is compared. Rows with an empty key are reported as added or removed.
The report is a CSV for Excel with one line per difference: status, key, row numbers in each file, column, before and after.
Tested on 300 random pairs of shuffled, edited tables against a brute-force comparison. Nothing is uploaded.
Select both files on Compare Two CSV Files, the older one first. Rows are matched on a key column chosen for you, such as id or email, and the counts of added, removed and changed rows appear at once, with every changed cell before and after. Download saves the list as a CSV.
Yes. Pick two .xlsx files, or one .xlsx and one CSV; the first sheet of each workbook with data is compared. Formulas are compared by their saved results. To compare another sheet, save it on its own as CSV from Excel and pick that file.
Order does not matter. Each row is found by its key value, so a list sorted by name and the same list sorted by date compare as identical. Without a key, choose Compare by row position, and row 1 is compared with row 1, row 2 with row 2.
Numbers are compared as numbers by default, so 1.50 and 1.5, or 1,000 and 1000, are the same value written differently. Turn off Numbers as numbers to compare the exact text instead. Spaces at the start and end of a cell are ignored unless you turn that off too.
When a key appears more than once in a file, the first row with it is compared and the duplicates are listed under the summary, so you can clean them up. Rows with an empty key cannot be matched and are listed as added or removed.
No. Both files are read and compared by your own browser, and the report is written on your device. Customer lists, price lists and payroll exports never reach a server, and nothing remains once the tab is closed.
Rates and rules change; check the source before you rely on a figure.
Still stuck? Ask about Compare Two CSV Files and anybody who knows can answer. No account needed.
On the related tools: Convert Any File – Photos, documents, spreadsheets, archives: drop it and see what is possible · Markdown File to HTML – Turn a .md file into the HTML behind it.