Compare Two CSV Files Online
Find records in File A that do not already exist in File B. Map matching columns, review the results and download a CSV directly in your browser.
CSV file comparison
Load both files to map their matching columns. File A is your incoming list; File B is your existing reference.
How CSV comparison works
A CSV comparison tool checks whether the records in one file are present in another. RowDesk uses the matching columns you choose, rather than assuming that the files share the same headers or column order.
File A is the incoming list. File B is the reference, such as an existing customer export. The tool separates new A records, matched A records and records found only in B. It compares keys, not every changed cell.
Find new records before import
Compare two CSV files without spreadsheet lookups or uploading their contents.
- Load both files
Open your incoming CSV as File A and the existing export as File B.
- Map the keys
Choose equivalent fields in both files. Set whitespace and capitalization rules deliberately.
- Review the results
Inspect counts, sample records and match explanations. Check incomplete matching fields.
- Download new rows
Choose Only in A, complete the first-export email step and download a separate CSV.
Compare the fields that identify a record
Map Email in A to Email Address in B to compare lead lists. For company-specific identifiers, you might map both Customer ID and Company. Every mapped pair must match for a row to appear in both files.
Use stable identifiers where possible. Names alone may match different people. Enable trimming for accidental surrounding spaces; enable ignoring capitalization only when case should not distinguish your keys.
Matching does not combine records or decide which copy is best. Use the exact duplicate CSV remover for repeated whole rows, or explore RowDesk duplicate matching and review for a broader workflow.
New, matched and only in B
Only in A contains A rows without a matching key in B. In both contains A rows with a key present in B. Only in B contains B rows without a matching key in A.
Repeated keys are not deduplicated or expanded into joined pairs. For example, two A rows matching one B row count as two matched A rows and one matched B row. The separate totals make this distinction visible.
To find differences between CSV files, first decide whether you need new records or a full value-by-value comparison. This utility answers the first question. The guide to comparing two CSV files explains how to choose useful keys and verify results.
Check incoming lists against existing records
Compare a campaign lead list with a CRM export before importing new contacts. Check an incoming customer file against your current customer list, or inspect which supplier IDs appear in only one export.
Review your destination system requirements before import. Comparison does not validate every field or upload records to a CRM. Use the free CSV cleaner for formatting cleanup, or connect preparation steps in the full RowDesk Compare workflow.
Compare locally. Keep your source files.
Both files are parsed and compared in a browser worker. Raw rows and cell values are not sent to a server or analytics by this utility. The email request is separate and contains no CSV contents.
Exports are generated locally from the selected result. Neither source file is overwritten, and closing this page clears its working data. Read about RowDesk browser-local processing.
CSV comparison, explained.
Can the two CSV files have different column names?
Yes. Map the identifying column in File A to its equivalent in File B, such as Email to Email Address. Column order and unrelated columns can differ. Add more pairs when a single field is not enough to identify a record.
Does comparison change my original values?
Trimming and case normalization affect matching keys only. Downloaded data retains its source values and order. Optional spreadsheet formula protection adds an apostrophe to formula-like cells and headers; the preview shows this output.
Why can matched counts differ between the files?
This tool counts source rows, not unique keys or every possible pair. If two A rows share a key found in one B row, both A rows count as matched, while one B row counts as matched. Each row appears once in its result.
What happens when a matching field is empty?
A row with any empty or whitespace-only matching field stays unmatched. It appears in Only in A or Only in B. Review these rows carefully: a missing identifier does not prove that the business record is new.
Does this find every changed cell between two files?
No. It finds which selected matching keys exist in each file. Non-key fields may differ even when two records match. It does not produce a cell-by-cell diff, merge values or detect fuzzy matches.
Do I need an account to export?
No account is required. The shared email gate asks for an email before your first export, with an optional unchecked marketing choice. Successful completion is remembered across all four free tools for 30 days in this browser.
What files can I compare?
Use UTF-8 CSV files, up to 25 MB each, with no more than 500 columns per file and a combined processing budget of 2 million cells. These safeguards apply to this lightweight utility, not RowDesk subscription pricing. Device memory can also limit processing.