How to identify conversations that need human review

    CategoryAI Employee
    Article typeDocumentation
    Estimated reading time 4 min
    Last reviewedRecently reviewed
    Applies toGEVADE CRM accounts where the feature is enabled
    Access requiredAccess may depend on your user permissions
    Plan/configuration noteFeature availability may depend on your selected plan and account configuration

    Summary

    Set up workflow automations to tag or flag conversations based on negative keywords, multiple fallback responses, or specific high-value intent triggers.

    Who this article is for

    This article is for support managers and administrators optimizing their team's workflow.

    Overview

    Not all AI conversations require human eyes, but missing the ones that do can cost you business. You can use automation to proactively flag conversations that might be going poorly, even if the customer hasn't explicitly asked for a human.

    Before you begin

    • Access Workflow Automation
    • Define keywords indicating frustration
    • Set up notification alerts

    Step 1: Step 1: Identify Review Triggers

    Determine what makes a conversation risky. This could be negative sentiment words ('frustrated', 'cancel', 'wrong'), or the AI using its fallback response ('I don't know') more than twice.

    Step 2: Step 2: Create a Monitoring Workflow

    Navigate to Automation > Workflows. Create a new workflow triggered by 'Customer Replied' or a specific AI action.

    Step 3: Step 3: Add Condition Branches

    Set a condition to check the message content for your negative keywords. If the condition is met, move to the next step.

    Step 4: Step 4: Tag and Notify

    If the condition is met, add an action to tag the contact 'Needs Review' and send an internal notification to the support team.

    Step 5: Step 5: Create a Smart List

    In the Contacts tab, create a Smart List filtered by the 'Needs Review' tag. Your team can check this list daily to audit flagged conversations.

    Configuration note

    We do not guarantee AI accuracy, lead conversion, bookings, customer satisfaction, response quality, revenue, support resolution, or business outcomes. The AI Employee does not fully replace human staff. AI behaviour depends on knowledge sources, instructions, handoff rules, enabled features, connected channels, permissions, configuration, testing, and ongoing monitoring. AI-generated responses should be reviewed before relying on them for sensitive, urgent, legal, financial, medical, billing, or policy matters. The AI Employee may need human handoff for these situations. Knowledge sources must be reviewed, updated, and tested regularly. Usage charges, feature access, fair use rules, or add-on requirements may apply depending on your account setup. Always test before going live. Contact support if an option is not visible. This proactive approach is different from a standard handoff. A handoff stops the AI; this monitoring approach flags the conversation for a human to review quietly in the background.

    Important clarification

    Setting the keyword triggers too broadly (e.g., triggering on the word 'no'). This will flood your team with false alarms. Use specific phrases like 'not helpful' or 'cancel account'.

    Resolution guidance

    My team is getting too many notifications.

    Refine your workflow conditions. Require a combination of factors, such as a negative keyword AND the contact being in a specific pipeline stage.

    Was this article helpful?

    Need account-specific help?

    If your question requires account access, billing review, workflow inspection, message delivery diagnosis, AI response review, or private configuration review, submit a support request with examples and screenshots.