How to Convert a Bank Statement to CSV
CSV is simple by design. That simplicity makes column choices, encoding, and verification matter more—not less.
Know what CSV preserves—and what it does not
A CSV file stores plain tabular values. It does not preserve workbook formulas, colors, multiple sheets, locked cells, or rich date formatting. That makes it portable across Excel, Google Sheets, accounting tools, and databases, but it also means every convention must be explicit in the columns themselves.
Use a header row with stable names such as date, description, debit, credit, amount, balance, and source_page. Save as UTF-8 so accented merchant names and non-English descriptions survive. Fields containing commas, quotation marks, or line breaks must be escaped correctly by a CSV writer rather than assembled by hand.
Extract, normalize, and keep an audit trail
Text-based PDFs may expose transaction text directly, while scanned statements need OCR. In either case, preserve the bank’s original transaction order and description. If you later create a cleaned merchant column, keep it beside—not instead of—the raw description.
Dates deserve special care. A statement may print only month and day on each row, leaving the year in the statement period. Transactions around December and January can therefore belong to different years on the same statement. Resolve the year from the period and row order; do not apply the download date to every transaction.
- Prefer ISO values such as 2026-08-12 when the destination expects machine-readable dates.
- Use decimal points and no currency symbols in numeric amount fields.
- Represent missing balances as empty fields, not zero.
- Preserve transaction order even if a user later sorts the file.
Test the CSV in its destination
Open the finished file in the system that will consume it. Excel and Power Query usually detect CSV delimiters and column types, but automatic type detection can still turn account-like identifiers into numbers or reinterpret ambiguous dates. Explicit import settings are safer than double-clicking a file when the locale is unknown.
Finally, total the signed activity and reconcile it to the opening and closing balances or printed credits and debits. A CSV can be syntactically perfect and financially incomplete. The source statement—not the number of exported rows—is the authority for whether the conversion is done.
Before you call it done
- UTF-8 encoding
- One header row
- One transaction per row
- Escaped descriptions
- Unambiguous dates
- Numeric amounts
- Empty means unavailable, not zero
- Balance reconciliation
Frequently asked questions
Is CSV better than Excel for bank statements?+
CSV is better for portable imports and simple automation. XLSX is better for human review, formatting, formulas, and multiple sheets. A strong workflow often provides both from the same normalized transaction data.
Why did Excel change my CSV dates?+
Excel may infer dates using your regional settings. Import the file through Data > From Text/CSV and choose the date type deliberately, or use unambiguous ISO-formatted values.
Sources and further reading
We prioritize regulators, public agencies, and first-party product documentation. Sources support the general guidance above; StatementMint’s workflow recommendations are our own.
- Import data from data sources with Power QueryMicrosoft Support
Educational information only—not financial, accounting, tax, or legal advice. Institution terms and your facts control.