Why a before-and-after report matters
A file can import successfully and still attach values to the wrong customers or SKUs. A trustworthy review needs to compare records by stable identifier, not only by row order or total dollar amount.
What a targeted verification report should show
- Original and candidate row counts, with additions and deletions called out.
- Every altered shared field linked to a stable record ID.
- Duplicate or blank IDs that make one-to-one matching ambiguous.
- Columns added or removed and numeric totals that remain unchanged despite row-level differences.
- Unresolved exceptions that require the owner to decide what is factually correct.
Synthetic counterexample: swapped customer contacts
| ID | Original email | Revised email |
|---|---|---|
| C101 | a@example.test | b@example.test |
| C102 | b@example.test | a@example.test |
Both rows still exist, and there are still two distinct emails—but the relationships are reversed. A set-of-values check would miss this; an ID-bound comparison exposes it.
Try the real demonstration
Our browser CSV comparison tool accepts local sample CSV files within its published limits and can download an exception list. The demo identifies differences; deciding which version reflects real-world truth requires supporting evidence.