Fix a bank statement CSV that opens in one Excel column

Check whether the data is still intact

A one-column display does not necessarily mean the CSV is damaged. The original record may still contain all its fields, separated by commas, semicolons or another character that Excel did not interpret as intended.

Preserve the original download and inspect a copy in a text editor. Compare the header with several transaction records. If the separators are present and the records are consistent, changing the import interpretation may be enough; extracting the PDF again would add an unnecessary step.

Read a fictional record before splitting it

Consider this illustrative comma-delimited record: 2026-09-04,"Sample supplies, downtown",-84.50. It contains three fields. The comma inside the quoted description belongs to that description; it is not a fourth column.

Now consider a semicolon-delimited record: 2026-09-04;Sample supplies;-84.50. Selecting a comma as the delimiter leaves it in one column. These examples show why the filename extension alone cannot tell you which separator the file actually uses.

Correct the preview, then inspect the types

In desktop Excel versions that provide it, use Data and From Text/CSV to import the file and choose its delimiter in the preview. Microsoft’s legacy Text Import Wizard also exposes delimiter and text-qualifier settings. Control availability depends on the installed version.

The preview should place dates, descriptions and amounts under their respective headers. Correct separation is only the first check: preserve identifiers as text, verify the date order, and confirm that negative amounts remain negative. Do not make a system-wide region change just to solve one file’s import.

Look for four warning patterns

Review several records, including descriptions containing punctuation. A preview of only the first simple row can conceal a problem later in the file.

• One field per row: the chosen delimiter may be absent from the file.
• Descriptions spread across columns: quoting or the text qualifier may be wrong.
• Amounts split at a comma: check whether the comma is a decimal separator or part of a quoted value.
• Rows have different apparent field counts: inspect embedded punctuation, line breaks or an incomplete export before deleting anything.

Validate the repaired import

Compare transaction count, first and last dates, and amount totals with the original records. Keep summary balances outside the transaction list. If a description spans a quoted line break, count parsed records rather than assuming every physical text line is a separate transaction.

Save a separate XLSX working copy once the import is correct. If the source itself is malformed or truncated, request a fresh export rather than making a series of unexplained splits. Fulla’s fictional sample CSV offers a known reference file for practicing the import checks.