About this article
This article was created using an automated generation workflow leveraging generative AI. We reviewed the Microsoft Learn Import-Csv -Delimiter specification and structured a way to test troubleshooting when a CSV appears as a single column using dummy data. No PowerShell execution was performed during the creation of this article.Verification status: 📘 Microsoft official specification checked / PowerShell execution unverified
Information verification date: October 2, 2026.
Even in a CSV, the delimiter is not necessarily a comma. Reading a semicolon-delimited file with the default comma can make it look like a single column.
Check the column count
@' Id;Name 1;Alice 2;Bob '@ | Set-Content -LiteralPath "$env:TEMP\demo.csv" -Encoding utf8 $data = Import-Csv "$env:TEMP\demo.csv" -Delimiter ';' $data[0].PSObject.Properties.Name
The default value for Delimiter is a comma. If the specified character does not match the actual delimiter, columns cannot be correctly separated into distinct properties.
Change one place
Import-Csv "$env:TEMP\demo.csv" -Delimiter ','
If the entire header looks like a single property name, you can suspect the Delimiter before jumping to the conclusion that the data is corrupted.
Verify column count programmatically
$row = Import-Csv "$env:TEMP\demo.csv" -Delimiter ';' | Select-Object -First 1 $row.PSObject.Properties.Count
flowchart TD A["CSV受領"] --> B["区切り文字確認"] B --> C["Import-Csv -Delimiter"] C --> D["列名/列数確認"] D --> E["後続処理"]
For professional use
Also check the character encoding, quotation marks, line breaks, and header names. Validate required columns immediately after loading, and halt subsequent processing if they are unexpected.
"A CSV showing as one column" does not necessarily mean it is corrupted
Even if called a CSV, actual files sometimes use delimiters other than commas. Rather than assuming Import-Csv is a universal parser that automatically detects formats visually, it is safer to verify the delimiter and header of the received data before loading.
Particularly in automated processes, cases where Import-Csv does not error out but instead "successfully reads the data as a single column" can be tricky. Validating the expected property names or column counts upfront rather than passing the data directly to subsequent steps prevents erroneous updates.
