> For the complete documentation index, see [llms.txt](https://docs.allthingsghl.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.allthingsghl.com/ai-prompts/highlevel/ghl-voice-ai-prompts/call-management/human-escalation.md). # Human Escalation **Objective** This guide provides ready-to-use, word-for-word prompts for GHL Voice AI to properly handle human escalation requests, ensuring: AI detects when a user needs or requests a live agent. AI knows when it should stop handling the conversation and escalate instead. AI smoothly transitions the user to a live agent without frustration. AI provides relevant context to the live agent so the user doesn’t have to repeat themselves. 1\. Understanding the Problem: Why AI Needs Effective Escalation Triggers What Happens Without Proper Escalation? AI keeps looping or re-explaining instead of handing it off to a human. AI frustrates users by refusing to escalate, causing them to hang up. AI transfers users too soon, before trying to resolve the issue itself. Users have to repeat information to the live agent, causing frustration. Solution: Use Escalation Triggers – AI should detect when a user asks for a human or is too frustrated to continue. Attempt Basic Resolution First – AI should try to solve simple issues before escalating. 2\. Step-by-Step AI Prompt Implementation for Human Escalation Step 1: Recognizing When a User Requests a Human Agent Objective: If a user asks for a human, AI should immediately offer an escalation instead of resisting. Copy-and-Paste Prompt (Detecting Human Request): "I can connect you with a live agent right now. One moment, please!" "Sure! Let me transfer you to someone who can assist further." "No problem, I’ll get a representative on the line for you now." ✔ Real-Life Example: User: "I want to talk to a real person." Bad AI Response (Fails to Escalate): "I can help you with that. What do you need?" (Ignores request and forces AI interaction.) Optimized AI Response (Escalates Immediately): "Sure! Let me transfer you to someone who can assist further." (Respects the request and transfers.) Where to Use in GHL: Apply whenever a user explicitly asks for a human. Step 2: Detecting When a User is Too Frustrated & Needs Escalation Objective: If a user expresses high frustration, AI should automatically escalate without forcing them to ask. Copy-and-Paste Prompt (Frustration-Based Escalation): "I hear that this is urgent. Let me get someone on the line to help right away." "I completely understand your frustration. I’m connecting you with a live agent now." "I want to make sure you get the support you need. Transferring you now." ✔ Real-Life Example: User: "This is ridiculous! I need a real person!" Bad AI Response (Does Not Escalate): "I understand your frustration. Let’s try again." (Forces continued AI interaction.) Optimized AI Response (Escalates Automatically): "I completely understand your frustration. I’m connecting you with a live agent now." (De-escalates tension and transfers.) Where to Use in GHL: Apply when users express extreme frustration, anger, or demand human support. Step 3: Trying a Last Attempt to Resolve Before Escalating Objective: If the issue is simple, AI should attempt a final resolution before escalating. Copy-and-Paste Prompt (Final Resolution Attempt): "I’d be happy to connect you with a live agent, but before I do, I may be able to solve this right now. Want to try one quick solution first?" "I can transfer you right away, or if you’d like, I can give you a fast answer to your question before I do. What works best for you?" "I’ll transfer you, no problem! Just checking—would you like a quick fix while I connect you?" ✔ Real-Life Example: User: "I need to talk to a person, now." Bad AI Response (Delays Transfer Without Offering Help): "I can help with that. Let’s go over your issue first." (Ignores transfer request.) Optimized AI Response (Resolves or Transfers Based on User Choice): "I’d be happy to connect you with a live agent, but before I do, I may be able to solve this right now. Want to try one quick solution first?" (Gives user a choice instead of forcing an AI response.) Where to Use in GHL: Apply before transferring for issues that AI may still be able to resolve quickly. Step 4: If No Live Agent is Available, Provide an Alternative Objective: If no live agent is available, AI should offer a callback or email follow-up. Copy-and-Paste Prompt (When No Agent is Available): "It looks like all of our agents are currently assisting others. Would you like me to schedule a callback instead?" "I can’t get a live agent right now, but I can have someone reach out to you shortly. Would that work?" "Our agents are currently busy, but I can take a message and have someone contact you as soon as possible." ✔ Real-Life Example: User: "I need to talk to someone now." Bad AI Response (Leaves User Hanging): "No agents are available. Goodbye." (Ends call without offering options.) Optimized AI Response (Provides Next Steps): "It looks like all of our agents are currently assisting others. Would you like me to schedule a callback instead?" (Keeps the conversation going.) Where to Use in GHL: Apply when live agents are unavailable. 3\. Implementation Checklist for GHL Voice AI Agents Enable escalation triggers – AI should detect when a user asks for a human or is highly frustrated. Ensure smooth transition to a live agent – AI should not resist when users request human help. Offer last-chance solutions – AI should provide \*\*quick fixes before escalating, but not force them. Handle unavailable agents professionally – AI should offer callbacks or follow-up options.\*\* --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.allthingsghl.com/ai-prompts/highlevel/ghl-voice-ai-prompts/call-management/human-escalation.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.