How to measure AI Employee performance after launch

    CategoryAI Employee
    Article typeDocumentation
    Estimated reading time 5 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

    Track key metrics like Resolution Rate (conversations completed without human handoff), Appointment Booking Rate, and AI Usage Costs via the dashboard.

    Who this article is for

    This article is for business owners and managers evaluating the ROI of their AI automation.

    Overview

    To justify the cost and effort of the AI Employee, you must track its performance. Measuring the right metrics tells you if it is actually saving your team time and generating revenue.

    Before you begin

    • Define your success metrics
    • Access the AI reporting dashboard
    • Compare AI performance against human baselines

    Step 1: Step 1: Track Resolution Rate

    This is the percentage of conversations the AI handles from start to finish without requiring a human handoff. A low resolution rate means your knowledge base needs work; a high rate means the AI is successfully deflecting support tickets.

    Step 2: Step 2: Track Appointment Booking Rate

    If your AI's objective is booking, measure how many conversations result in a scheduled calendar event. Compare this to your previous manual booking rate.

    Step 3: Step 3: Monitor Escalation Volume

    Track how many conversations trigger your human handoff rules. A sudden spike in escalations usually indicates a broken integration, an outdated FAQ, or a new customer issue the AI hasn't been trained on.

    Step 4: Step 4: Review Usage Costs

    The AI Employee consumes credits. Review your billing dashboard to track the cost per conversation. Ensure the time saved by your staff outweighs the AI usage cost.

    Step 5: Step 5: Conduct Qualitative Reviews

    Metrics don't tell the whole story. Continue to perform weekly manual reviews of conversation transcripts to gauge customer sentiment and tone accuracy.

    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. Do not aim for a 100% Resolution Rate. A healthy AI setup should escalate complex issues. A 100% rate usually means your escalation triggers are broken, leaving customers trapped with an unhelpful bot.

    Important clarification

    Focusing solely on the number of messages sent rather than outcomes. 50 messages sent to book one appointment is less efficient than 5 messages sent to book one appointment.

    Resolution guidance

    My booking rate is low.

    Review the AI's qualification questions. If it asks too many questions before offering a calendar link, customers will abandon the chat. Streamline the booking flow.

    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.