A CSV file (Comma-Separated Values) is a plain-text file that stores data in rows and columns — like a simple spreadsheet saved as text. It is the most common format used by banks, PayPal, accounting software, and online services to export your transaction history.

Because CSV has no single enforced standard, every bank and every application creates its own flavour. This converter automatically detects most differences, but understanding them helps if something does not look right.

Field Separators — the "comma" that may not be a comma

The character that separates one value from the next is called the delimiter. Despite the name "Comma-Separated Values", many files use a different character:

  • Comma (,) — the most common choice in English-speaking countries and the US.
  • Semicolon (;) — widely used in Europe (France, Germany, Spain, Netherlands). This is because European countries write numbers with a comma as the decimal point (e.g. 1.234,56), so the comma cannot also be used to separate fields.
  • Tab — common in database exports and files saved from Excel as "Tab-delimited text". These files sometimes have a .tsv extension instead of .csv.
  • Pipe (|) — used by some older accounting and banking systems.

This converter detects the separator automatically. If columns appear merged or shifted, check whether your file uses a semicolon or tab instead of a comma.

Date Formats — the source of most conversion problems

CSV files store dates as plain text with no label saying which part is the day and which is the month. This means 01/05/2024 could be the 1st of May or the 5th of January, depending on who created the file.

  • DD/MM/YYYY (e.g. 25/12/2024) — United Kingdom, Australia, most of Europe.
  • MM/DD/YYYY (e.g. 12/25/2024) — United States.
  • YYYY-MM-DD (e.g. 2024-12-25) — International standard (ISO 8601), databases, modern APIs.
  • DD MMM YYYY (e.g. 25 Dec 2024) — Some bank statements and PDF exports.
  • Short years (e.g. 25/12/24) — Two-digit years add further ambiguity.

This converter scans your file and detects the date format automatically. If dates appear incorrect in the output, use the country selector on the next step to override the detection.

Quoting — values that contain commas

If a field value itself contains a comma (for example, a description like "Food, drinks and tips"), the entire value is wrapped in double quotes to prevent the comma from being read as a separator:

"Food, drinks and tips",25.00,2024-01-15

Most files do this correctly, but some older exports skip the quotes, or use a backslash instead of doubling a quote character. The converter handles all common quoting styles.

Line Endings — invisible characters that differ by operating system

Every row in a CSV ends with an invisible end-of-line character. Three variants exist:

  • Windows (CRLF) — two characters: carriage return + line feed.
  • Linux / Mac (LF) — a single line-feed character. The most common on modern systems.
  • Old Mac (CR) — a single carriage-return, used by classic Mac OS before 2001. Very rare today.

This converter handles all three variants and mixed files automatically.

BOM — the invisible character that can break the first column

A BOM (Byte Order Mark) is a hidden three-byte signature that some Windows applications (particularly older versions of Microsoft Excel) add to the very beginning of a file to identify the encoding. On Unix and Mac systems it is not used and is considered an error.

The BOM creates a phantom invisible character at the start of the first column header. This can cause the first column name to not be recognised correctly. This converter detects and removes the BOM automatically before processing your file.

Character Encoding — why special characters sometimes appear garbled

Computers store text as numbers. "Character encoding" is the rulebook that maps those numbers to letters. Two files using different rulebooks will corrupt each other's special characters.

  • UTF-8 — the modern standard, used by most websites and modern software. Handles every language and symbol.
  • Windows-1252 (ANSI) — an older Windows encoding. Handles Western European characters (£, €, é, ü) but not characters from other alphabets. Many older bank export systems use this.
  • ISO-8859-1 (Latin-1) — similar to Windows-1252. Common in older European banking software.

The most visible symptom of an encoding mismatch is garbled currency symbols: £ appearing as £, or € appearing as €. This converter automatically detects your file's encoding and converts it to UTF-8 before processing.

Why bank CSV files are all different

Unlike OFX and QIF — which are purpose-designed financial formats with strict rules — CSV has no banking standard. Each bank, payment processor, and accounting package invents its own column names, date format, and amount layout. Barclays, NatWest, PayPal, and American Express each produce completely different CSV structures. This is why a mapping step is needed: you tell the converter which column in your file contains the date, which contains the amount, and so on.

Accepted file types

  • .csv — standard comma-separated values file.
  • .tsv — tab-separated values file (the same format, just with tabs instead of commas).

Maximum file size and row limits depend on your account type. Registered users can process larger files.

Select the CSV file (Step 1 of 5)

Drop a file here or click to upload.

CSV TSV