By key column with CSV Reconcile
This opens two small product exports, last week and this week, matched on the sku column. Two rows changed price or stock, one row was added and one removed. Replace either file with your own, pick the key column and press Compare.
- Choose a file for each box, drop one on it, or paste the text. The first row must be the header.
- Pick the key column that identifies a row. One is suggested if a column is unique and named like an ID.
- Open Comparison rules if whitespace, case or number formats should not count as changes.
- Press Compare and read the Changed, Added and Removed tabs.
- Download the changed rows or every difference as CSV.
What to know
A line-by-line diff answers a different question from the one you have. If the second export is sorted differently, or someone added a column in the middle, nearly every line differs and the real changes are buried. Matching on a key turns the problem around: each record in one file is looked up by its ID in the other, and only then are the cells compared. Row order and column order stop mattering.
In the sample, six products are compared. Two keep every value, so they are counted as unchanged. A-100 lost 4 units of stock, A-101 went from 59.00 to 64.00, and A-104 dropped from 9 to 3. A-102 exists only in the first file and A-106 only in the second. The Columns tab adds the arithmetic: stock changed in two rows by 10 units in total, price in one row by 5. The updated column holds a different date on every row, so the sample ignores it; otherwise every row would count as changed.
Choose the key carefully. A good key is unique in both files and does not itself change between exports. A name is a poor key, an ID or SKU a good one. If the key you pick repeats, those rows are not compared at all; they are listed under Ambiguous so that nothing is silently matched to the wrong record.
Updated · Cosmovex
Questions
Does the order of the rows matter?
No. Rows are matched on the key, not by position, so a re-sorted export shows no changes. The same holds for the order of the columns, because columns are matched by name.
What if my files have no ID column?
Select up to three columns together as a combined key, for example first name, last name and birth date. If no combination is unique, the repeated rows are reported as ambiguous rather than compared.
Are my files uploaded?
No. Both files are decoded, parsed and compared in your browser tab and are not sent to a server.
Can I compare tab or semicolon separated files?
Yes. The separator is detected from the first lines: comma, semicolon, tab or pipe. The page shows which one it used for each file.
How do I get the result into a spreadsheet?
Download the changed rows or all differences as CSV and open them in Excel or Google Sheets. Pro also writes an Excel report with one sheet per result type.
The free plan covers everything on this page. CSV Reconcile Pro ($29, paid once) is described on the CSV Reconcile page.
