Compare two messy files
Find missing rows, duplicates, changed values and almost-matches.
Step 1: Load both files
How Compare works
Drop two files. MessyMatch parses them in your browser, normalises values with the cleaning rules you pick (trim, casing, accents, leading zeros, invisible characters) and produces an indexed diff in seconds. Originals are never modified. The normalised form is used only for matching.
- Load file A and file B (CSV, TSV, TXT, XLSX or pasted text).
- Choose full-row or key-column comparison and the cleaning rules to apply.
- Optionally enable fuzzy matching to surface near misses with a reason (accent, casing, typo, word reorder).
- Click Compare. The engine builds a hash index and walks both files in parallel.
- Review the seven sections: only in A, only in B, in both, duplicates in A, duplicates in B, changed rows, almost-matches. Export any of them as CSV or XLSX.
When people compare two files
Compare exists for the cases where Excel formulas fall over: messy values, large files, two systems that should agree and do not.
Reconciling two exports of the same dataset
A CRM export and an ERP export of the same customers, two database snapshots, last week vs this week. Find what moved, what disappeared and what is duplicated.
Migrating data between systems
Check every row landed. Drop the source export and the target export, compare by primary key, fix the missing rows before going live.
Reconciling lists with messy text
Customer names with extra spaces, IDs with leading zeros, emails with mixed casing. Cleaning rules normalise both sides before matching.
Finding duplicates inside one file
Compare a file against itself by key column. Repeated keys are reported separately so you can dedupe before re-importing.
VLOOKUP and XLOOKUP replacement
Two columns, two files, no formulas. The diff is faster to read than a column of N/A cells and it catches almost-matches that VLOOKUP silently misses.
Compare FAQ
Are my files uploaded?+
No. Parsing and comparison run in your browser inside a Web Worker. The server only sees metadata (row count, file size, format) used to enforce abuse limits.
What formats are supported?+
CSV, TSV, TXT, XLSX and pasted text. Excel files above about 100,000 rows can be slow because XLSX is a zipped XML format. For very large files, export to CSV first.
What is full-row vs key-column comparison?+
Full-row treats the entire row as the matching unit. Key-column matches on the column you pick and reports changes in the other columns. Pick key-column whenever rows have a stable identifier.
What counts as an almost-match?+
Two rows that share a key but differ in another field after normalisation, or two rows whose keys are close (typo, accent, word reorder). Each almost-match carries a reason so you can decide what to do with it.
What does it cost?+
Free for files up to 2,000 rows each. Above that, the comparison is priced by the largest of the two files: $3 up to 25k, $7 up to 100k, $15 up to 500k, $29 up to 1M. One-off, no subscription.
Can I compare files with different column orders?+
Yes. In key-column mode the columns are matched by name. You can also map columns explicitly if the names differ between the two files.