Remove duplicate rows when merging CSV files
Overlapping exports are the usual cause of repeated rows: you download January to March, then February to April, and February appears twice. The tool can strip those repeats, but it treats the option with care because a repeated row is not always a mistake.
Your file is never uploaded. The PDF is read in your browser and makes no network requests while it works.
Open the Merge & clean CSV tool
Steps
- Add the overlapping CSV files to the tool.
- Find the Remove duplicate rows checkbox. It starts unticked.
- Tick it only if you are sure repeats in your data are copies, not separate events.
- Read the summary line, which reports N duplicate(s) removed, including when the number is 0.
- Compare the row count with what you expected before you download.
Good to know
Two identical bank rows can both be real
Buying the same coffee twice on one day produces two rows with the same date, description and amount. With duplicate removal on, one of them would vanish and your totals would be off. That is why the option is off unless you turn it on.
Only full matches are dropped
A row is removed only when every cell equals an earlier row after cleaning. One different character in any column keeps it. The first copy is always kept.
Clean first, then compare
Because trimming and tidying run before the comparison, rows that differ only by stray spaces count as matches.
Questions
Can it remove duplicates based on one column, like an ID?
No. It compares whole rows only, which is the safer behaviour for financial exports.
Why is it off by default?
Silently dropping a genuine repeated purchase is worse than leaving a true copy in place, which you can spot and delete yourself.
Starting from bank statement PDFs? RowSmith converts text-based PDF statements to CSV or Excel in your browser.
This guide was written with AI assistance and reviewed against the tool's behaviour. Always review your output before using it.