# Rehearse a difficult customer conversation Designed for: People preparing for a complaint or a difficult client call Result: A short practice conversation and feedback tied to an approved response policy. ## Try a sample first Illustrative example, not a recorded Bot run. This adaptation has not been runtime-tested. ### Complete first request Role-play a dissatisfied customer one turn at a time using the supplied scenario and policy. Afterwards, assess listening, factual accuracy and promises. Do not contact the customer or invent authority to offer compensation Use only the records I supply. Label missing or uncertain information. Do not connect accounts, browse, send, submit, publish, spend money or create a schedule. Return a draft for my review. Fictional scenario: a delivery is late. Policy P1: staff may check status and escalate; only a manager may approve refunds. Practice answer A1: 'I understand the delay is frustrating. I'll check the status. I guarantee a full refund today.' ### Example result A1 acknowledges the concern and offers a permitted status check. The guaranteed refund exceeds P1 and should be removed. Try: 'I understand the delay is frustrating. I'll check the status and ask a manager to review your refund request.' Next, practise responding when the customer asks for a firm arrival date you do not have. ### Check the answer - Tie feedback to P1 and A1. - Flag the unauthorized refund promise. - Do not contact a real customer or promise a result. Inspired by a community use shared by @FirstDeployAI: https://x.com/FirstDeployAI/status/2109048978759098826 Found through Bot Usecases (https://botusecases.netlify.app/), curated by KP. This fictional sample was written by Bot Cabinet. ## Operating guide - **When to use it:** Use this workflow when you need a short practice conversation and feedback tied to an approved response policy. - **Lead Bot:** Call Coach - **Cadence:** When a new request arrives and a person assigns it - **Typical first run:** 15–30 minutes for a first manual run ### Access for the first run - The approved inputs listed on this page - Only the files, sources, and services required for this run - No sending, publishing, spending, or live-system changes without approval ## Recommended starting Bot 1. Call Coach Use Call Coach with the task below. This is a workflow for an existing role, not a separate installed Bot. ## Information to gather - Start with the fictional records below - For your own task, choose only the records you want the Bot to use - Remove private details you do not need for the task ## Workflow ### 1. Call Coach Role-play a dissatisfied customer one turn at a time using the supplied scenario and policy. Afterwards, assess listening, factual accuracy and promises. Do not contact the customer or invent authority to offer compensation Output: A short practice conversation and feedback tied to an approved response policy. Message to send: Use the approved inputs I provide. Role-play a dissatisfied customer one turn at a time using the supplied scenario and policy. Afterwards, assess listening, factual accuracy and promises. Do not contact the customer or invent authority to offer compensation. Return this result: A short practice conversation and feedback tied to an approved response policy.. Ask me about missing information before you continue. ## Overall request Role-play a dissatisfied customer one turn at a time using the supplied scenario and policy. Afterwards, assess listening, factual accuracy and promises. Do not contact the customer or invent authority to offer compensation Use only the records I supply. Label missing or uncertain information. Do not connect accounts, browse, send, submit, publish, spend money or create a schedule. Return a draft for my review. ## Decisions for a person - Check the result against the supplied records - Approve any account connection, schedule or outside action separately - Review anything before it is sent, submitted, purchased or published ## First test Tie feedback to P1 and A1. Flag the unauthorized refund promise. Do not contact a real customer or promise a result. ## Success checkpoint Check the result against your original material. You make the decisions listed below. Tie feedback to P1 and A1. Flag the unauthorized refund promise. Do not contact a real customer or promise a result. ## If the workflow stalls If the result is incomplete, mark the missing information and ask the same Bot to correct it. Add another Bot only when a genuinely separate role or independent check would help. ## Hermes Desktop setup 1. Open https://botcabinet.com/bots/call-coach/. Copy its role instructions into a new Bot and review its limits. This specialist has no Hermes profile archive. 2. Keep outside accounts and schedules disconnected for the first test. 3. Run the complete sample task in its chat and check the answer against the supplied records. 4. Approve any additional access or outside action separately.