In this scenario, we automate the processing of incoming emails from GMAIL, determine their category using artificial intelligence, and forward them to operators for further handling. This will help to quickly and efficiently process customer inquiries without the need for intervention at the initial stage.
Example of an implemented scenario
Stages of building the scenario:
Stage 1: Receiving an email through the Action block receive_email_message
More details
receive_email_message
block automatically reads all new emails received in the specified GMAIL mailbox.email_message
variable, and the email subject in the email_subject
variable.Stage 2: Categorizing the inquiry topic using the Action fastline_completion
fastline_completion
block uses artificial intelligence to analyze the email text and determine the best inquiry category.email_message
) and its subject (email_subject
) and determines which category the inquiry belongs to. For example, it could be a consultation, complaint, or clarification.completion
variable.{
"model": "gpt-4o",
"temperature": 0.3,
"prompt": "You need to categorize customer emails. Based on the provided message '{{email_message}}' and subject '{{email_subject}}', determine the most appropriate topic from the following: consultation, complaint, clarification. Return only one topic without additional explanations.\\\\nImportant: choose only one category. If none of them fit, return 'Other' (other)."
}
Stage 3: Transferring the result to the operator panel using the Action operator_panel__connect_to_operator_with_msg