Prompt Optimizer for Conversations AI - Test, Tune, and Launch Better Chat Agents

You’ve developed a Conversations AI agent — but will it manage real customer chats in the way you anticipate? The Prompt Optimizer now integrates with Conversations AI, allowing you to assess its performance before your customers do.

It generates realistic customer scenarios, conducts actual chat conversations with a cloned version of your agent, evaluates the outcomes, identifies any shortcomings, and rewrites the prompt to enhance its performance. Your production agent remains unchanged until you review and implement the improved version.

Configure – Build Your Test Plan

Auto-generated scenarios: The AI creates contextual test scenarios based on your agent’s prompt, language, knowledge base, appointment setup, and actions. Each scenario comes with a customer persona, an opening message, expected behaviours, and a priority level.
Reuse past scenarios: Load scenarios from earlier runs instead of starting from scratch.
Multi-language support: Scenarios, evaluations, and optimised prompts remain consistent with your agent’s language.
Usage visibility: Monitor daily free messages used versus remaining, along with a pre-run checklist before you start.

Test – Real Chats, Real Actions

Real conversations, not simulations: The Prompt Optimizer sends genuine customer-style messages and waits for your agent’s actual responses.
Action tracking: Observe when the agent triggers Appointment Booking, Human Handover, Workflow Triggers, Bot Transfer, Stop Bot, Auto Follow-up, Contact Field Updates, and Knowledge Base Queries.
Full transcripts with AI scoring: Each chat is assessed against expected outcomes, providing clear reasoning for what succeeded, what failed, and why.
Runs in the background: Leave the screen and return to completed results.

Improve – AI-Guided Prompt Optimisation

One-click improvise: The AI analyses failed chats, identifies root causes, generates an improved prompt, and tests it.
Auto optimise: Execute multiple optimisation attempts automatically until your target accuracy is reached.
Prompt diff viewer: View precisely what changed before applying any modifications.
Best variation tag: The highest-accuracy prompt is highlighted, along with a full history of every attempt.

Testing runs occur against a cloned agent — your live agent is not altered until you click “Use Prompt.”
Temporary test contacts are created for each run and are automatically cleaned up, ensuring your CRM stays free of test data.

AI evaluations may vary between runs — consider accuracy scores as directional guidance.
Real actions (bookings, workflows, handovers) execute during testing, so use test calendars and workflows where necessary.
Long or unresponsive conversations time out automatically, ensuring runs always complete.