Why CSVs need a keyed compare
Export the same table twice and the rows often come out in a different order. A line-by-line diff then reports almost everything as changed. Matching rows by a key column fixes that: in the example, rows are shuffled, yet the result is exactly one price change, one stock change, one discontinued product and one new one.
Pick a column whose values are unique per row: an ID, SKU, order number or email. If the files have no such column, choose (whole row) to see which complete rows appear in only one file, or use compare lists for a single column.
Questions
How do I compare two CSV files?
Open or paste both files, then choose the key column that identifies a row (for example id, SKU or email). Rows are matched by that key, whatever order they are in, and the result lists added rows, removed rows and every changed cell.
What if my CSV has no ID column?
Choose "(whole row)" as the key. Each row is then treated as a single value, and you get rows only in the first file and rows only in the second file.
Does it support semicolon or tab separated files?
Yes. The delimiter (comma, semicolon, tab or pipe) is detected automatically for each file, and quoted fields with commas, quotes or line breaks inside them are parsed correctly.
Can I compare two Excel sheets?
Save each sheet as CSV (File > Save As > CSV) and open them here, or copy the cells and paste them: spreadsheets paste as tab-separated text, which is detected automatically.
How big can the files be?
Tens of thousands of rows compare in well under a second in the browser. The table shows the first 1,000 differences; the counts always cover every row.