CSV Reconciliation
Compare two CSV exports row by row using an exact or composite key. See what's unchanged, changed, missing or new, with every duplicate ID kept in view for you to decide. Free, private and runs entirely in your browser.
How to use CSV Reconciliation
- Load both UTF-8 CSV exports, choosing each file's delimiter first.
- Pick the key that identifies a row, such as SKU, or SKU plus warehouse, and map it across the two files' header names. Decide on trimming and ignore-case deliberately.
- Select the fields to compare. Numeric fields can use an absolute tolerance in the field's own units.
- Compare, check the counts, and review duplicate groups one row at a time, pairing rows yourself.
- Download the detail report, exceptions, the pair index and clean subsets, and save the rules as a recipe for the next run.
What people use it for
- Checking which SKUs changed quantity at each warehouse between two inventory exports
- Comparing a customer or contact list before and after a migration to see what was added, dropped or changed
- Verifying that an export from one system matches the import in another
- Spotting duplicate IDs that a plain spreadsheet lookup would silently hide
- Re-running the same monthly comparison with a saved recipe
Frequently asked questions
Is CSV Reconciliation really free?
Yes. It's free to use here in your browser and free to download. There's no account, trial or watermark.
Are my files uploaded anywhere?
No. The tool runs entirely in your browser and the page blocks it from making any network connection. Your files are processed in memory on your device; reloading the page clears them.
How does it decide which rows match?
By the key you choose: one column or a composite of up to five, compared as exact text, optionally trimmed and lowercased. There's no fuzzy matching and no numeric conversion of keys, so matches are predictable.
What happens with duplicate IDs?
A key that repeats on either side always goes to a review group instead of being matched automatically. You pair rows yourself, one at a time, and every decision is recorded in the pair index and reports. Undo returns the whole group to unresolved.
Can it ignore small number differences?
Yes. Numeric comparison fields accept an absolute tolerance, so 5.005 and 5.00 match at a tolerance of 0.01. Numbers must be plain signed decimals without thousands separators, currency symbols or exponents.
How large can the files be?
Up to 10 MiB, 50,000 data rows, 100 columns and 1,000,000 cells per CSV, with up to 5 key components and 30 comparison fields. It has been tested with 50,000 rows per side in Chrome on Windows.
Does it read Excel files?
No. Save each sheet as "CSV UTF-8" first. Files in UTF-16 or with invalid characters are rejected with a message so you can re-save them.
Is this accounting reconciliation?
No. It compares rows and fields under the rules you set. It doesn't do journal matching, payment splitting or fuzzy matching, and it can't prove two rows are the same real-world item.
Can I use it offline?
Yes. Click Download to save the tool as a single HTML file, then open it in any current desktop browser, with or without an internet connection.
Private by design
This tool is a single HTML file that runs entirely in your browser. The page that hosts it blocks every outgoing network connection from the tool, so the files you open are processed on your device and are never sent to AppKatz or anyone else. Download it once and it keeps working offline.
Take it further with AppKatz
Affordable Windows apps for small businesses. Buy once, own forever, no subscriptions.