BANK CSV & TABLE CLEANUP GUIDES

Prepare a bank CSV for import: dates, signs and duplicates

By Arjun Varma · Updated 19 September 2026

A bank CSV needs the right columns, date order and amount signs before you import it. Keep the original file, check how money in and money out are represented, then review a small output sample in your accounting software.

Try bank CSV prep — no signup →

A worked example: two amount columns become one

Here is the fictional example built into SnapCommit. The source uses semicolons between columns, day/month/year dates and decimal commas. Its column headings explicitly identify money coming in and money going out.

In bank CSV prep, click Try a sample. The fictional file loads with its known format and a review result. For your own file, select the formats explicitly; a date such as 01/09/2026 cannot tell you its date order by itself.

The review shows 1250.00 in, 84.40 out and 1165.60 net movement. Both software rows remain in those totals. One is flagged as a possible duplicate because the date, description and amount match. Check the original statement before deciding whether either row should be removed.

What the output looks like

With DD/MM/YYYY selected for output, the transaction CSV has these three columns. Money out is negative; money in is positive. Amounts use a decimal point.

Five checks before importing your file

1. Separate structure from number formatting. Confirm the CSV separator and header row first. A decimal comma is part of an amount, not necessarily a column boundary.
2. Map amounts by meaning. For one amount column, choose whether positive values mean money in or money out. For split columns, identify each side explicitly. Do not infer the meaning from “debit” or “credit” alone.
3. Choose the source date order. Confirm day/month/year, month/day/year or year/month/day against the original. SnapCommit requires four-digit years and rejects invalid calendar dates.
4. Resolve blocked rows and review warnings. SnapCommit blocks transaction export when a row is invalid. Possible duplicates remain for review. Its totals cover valid rows and do not reconcile your account or find transactions already in your ledger.
5. Preview the destination import. Choose the intended account, map the output fields, check date interpretation and compare signs and totals with the source before completing the import. Preserve the untouched original.

QuickBooks and Xero: check the destination requirements

Intuit’s CSV preparation guide describes a Date, Description, Amount layout with negative payments and positive deposits, as well as a four-column alternative. It also lists destination-specific formatting restrictions. A three-column file alone does not prove an import will succeed.

For Xero, consult its CSV mapping and date-format guidance. SnapCommit’s bank tool is beta: actual QuickBooks and Xero imports have not yet been verified. It is independent of both products. Review any small test import inside your own software.

Reuse the setup next month

After reviewing a file, choose Save setup for next month. The JSON file stores column headings, mappings and format choices, without transaction rows. Open the next export, read its columns and load the setup. The headings must match exactly; review the field examples again if the bank changes its export.

What is free?

The web tool checks up to 15 transactions without signup. The current $29 USD one-time offline release, plus tax, includes bank CSV prep for up to 2,000 transactions and a pasted-table cleaner for one person. Both process data locally. See the full offer and limits.

This workflow accepts CSV text, not PDF, image or Excel files. It does not categorize transactions, connect to a bank or convert currencies. Use one account and currency per file.

Related guides

• Clean pasted tables for Excel
• US and European number formats in copied tables
• Keep leading zeros and long IDs when pasting into Excel