How to use CSV Compare
- Open the old CSV in Before and the new CSV in After. Select their shared delimiter.
- Enter the exact name of a unique key column, such as sku. Leave it blank only when row positions define identity.
- Select Compare CSV. Review row counts, changed cells and column changes, then download the JSON report.
Reconcile two inventory exports
With sku as the key: 1 changed row, 1 added row, 1 removed row and 1 unchanged row.
Load this exact example ↗What changes—and what stays
All cells are strings. Column order is ignored because headers identify fields. A unique key matches records across row reordering; key values are exact and case-sensitive. Empty and duplicate keys are rejected. Optional cell trimming ignores outer whitespace in comparisons but preserves original values in the report.
Practical limits
Up to 1 MiB, 10,000 data rows and 200 columns per input; at most 5,000 reported differences. Both files need unique, non-empty headers. A single key column is supported.
Common questions
Can the two exports use a different row order?
Yes, when you choose a unique key. Without a key, the first data row is compared with the first data row, and so on.
What happens if a column is added?
The report lists added and removed headers. Matched rows also report the corresponding added or removed cells. Even a header-only file exposes column changes.
Does 00123 equal 123?
No. CSV values remain strings, so leading zeros are meaningful. Trimming only ignores outer whitespace; it does not convert numbers or lowercase identifiers.