PowerShell Write-Host vs. Write-Output

In PowerShell, you can output text with the Write-Host cmdlet. However, it is sometimes discouraged to use it. There is the alternative Write-Output, for example. In this post, I will explain the differences.

🎬 I also created a video on this topic.

Write-Host

Write-Host writes data to the PowerShell host, the program hosting the current PowerShell session. Often this is the ConsoleHost, but for example PowerShell ISE or Visual Studio Code use their own hosts, which differ from each other in some aspects.

Write-Host can influence the way data is output by the PowerShell host. For example, -ForegroundColor can be used to change the text color and -BackgroundColor to change the background color. So if you run a script interactively in a console, Write-Host can be quite useful for producing a more appealing output. One major disadvantage is that data written with Write-Host cannot be processed further in the pipeline.

So the result cannot, for example, be stored in a variable:

1
2
3
4
5

$Variable = Write-Host "Hallo PowerShell"
# Return: Hallo PowerShell

$Variable
# = no visible return

And it also cannot be piped to other cmdlets:

1
2

Write-Host "C:\temp" | Get-ChildItem
# Return: C:\temp (and not the contents of C:\temp!)

Write-Output

Write-Output, on the other hand, can be processed further:

1
2
3
4
5
6

$Variable = Write-Output "Hallo PowerShell"
$Variable
# Return: Hallo PowerShell

Write-Output "C:\temp" | Get-ChildItem
# Return: < contents of C:\temp ... >

For that, however, you lose the formatting options that Write-Host offers.

When to use which one?

It is beneficial to know and use the differences between Write-Host and Write-Output. For example, if you store the output of a scriptblock from a loop in a variable, outputs produced with Write-Output would likely pollute your result array with unwanted entries. If you use Write-Host instead, you can create output without it ending up in the result array.

1
2
3
4
5
6

$Ps1Files = Get-ChildItem *.ps1 -Recurse
$Ergebnis = foreach($File in $Ps1Files) {
    $File.FullName # This is output from the loop
    Write-Host "$($File.Name)" -NoNewLine -ForeGroundColor Yellow
    Write-Host " found..."
}

Or, if you run a script in Azure Automation, outputs created with Write-Host do not end up in the logs. If you create scripts for Azure Automation, you should therefore use Write-Output instead of Write-Host.

1
2
3
4
5

# Azure Automation example code
Write-Output "Azure Automation also uses another PowerShell host"
Get-Host

Write-Host "Output produced with Write-Host is not visible in Azure Automation"

Write-Output is actually optional

If you think that Write-Output should generally be preferred: you do not actually need to use Write-Output at all. You can simply reference the object you want to output directly, whether it comes from a variable, a cmdlet, or something else, and it will be displayed. The only advantage is that it is a bit more obvious in the code that you deliberately want to generate output.

1
2
3
4
5

Write-Output "Hello :)"
# Return: Hello :)

"Hello :)"
# Return: Hello :)

Note on output streams

Apart from Write-Output and Write-Host, there are also other output cmdlets, such as Write-Information, Write-Error, Write-Warning, and so on. You can use them to produce output and send it to different output streams. The person running the script at the end can decide which information, meaning which streams, they want to see. Interestingly, Write-Host does not only write directly to the console host, but also to the information stream. If you want to know more about this topic, check out this video about PowerShell output streams.

Conclusion

In summary, I would say: Write-Host mainly displays information visually for the user in the console. Write-Output is independent of that and passes information through the PowerShell pipeline. If Write-Output is the last command in a pipeline, the information is displayed visibly and can be captured better by other logging functionality.