Skip to main content

I'm creating a dummy workflow currently to test out AI phone call automation; I’ve gotten everything else working properly other than the conversation analysis. I make sure to keep the version in gpt-4o but every time I try and test it sends me back a failed notification. Any help would be greatly appreciated.

 

This post has been edited by a moderator to remove Zap links and personally identifiable information (PII). Please remember that this is a public forum and avoid sharing personal or potentially sensitive details.

Hey ​@Seth Peterson,

Please post the screenshot of the error from the test tab for us to have more context. Try increasing the max tokens and try changing your model to see if it fixes your error. Then click on the retest to see if it works. Hope it helps!


Try splitting your transcript into smaller chunks before analysis — long text inputs often cause failures. Also double-check that your workflow passes the text as plain input, not as a file object or encoded string.