Articles on: AI Setup & Follow-Ups

How to use Automatic Human Escalation to Hand Off Leads from the AI to your team

📝 Set rules so the AI pauses and hands off to your team when a lead is ready to book, asks for a person, or hits any criteria you define.



Table of contents




Overview



Human escalation lets you set rules and criteria that, when met, tell the AI to stop and hand the conversation to your team. When escalation fires, the AI either sends one final message or stops replying entirely — then a team member can step in to respond manually or resume the AI.


Where to find human escalation


You'll find all the settings under AI Knowledge Base → Human escalation.


From here you can set your sensitivity, define when to escalate, add example triggers, configure the messages the AI sends, and test everything in real time.



Set your sensitivity


Sensitivity controls how eager the AI is to escalate a conversation.


  • Low — least likely to escalate. The AI keeps working unless it's very clear escalation is needed. Good place to start.
  • Moderate — balanced. The AI escalates when it thinks it needs to.
  • High — most likely to escalate. If the AI is uncertain or the topic is sensitive, it escalates. Use this if you want most things handed to your team.



Start on Low to get comfortable, then dial up to High if you'd rather most conversations get escalated to a person.



Tell the AI when to escalate


The When to escalate field is a simple text prompt where you describe exactly when and how you want the AI to hand off. Keep it plain and specific.


Example: "Once the lead clearly indicates that they're ready for us to come out and service them, go ahead and escalate."


You can write whatever prompt you like, or use the LeadTruffle assistant to help you draft it.


Example triggers


Below the prompt you'll find a set of ready-made examples you can turn on or off. Common ones include:


  • The customer asks for a person
  • It's a warranty or prior work
  • They ask for someone to text them
  • The lead says they want to talk to a manager


Add one to include it, or click the X to remove it.




Additional message settings


You control what the AI says (if anything) when escalation fires.


Final acknowledgement


Choose whether the AI sends one last message before it stops. For example, if a lead asks to talk to a manager, the AI can reply:


"Thank you for your message. Someone from the team will be following up shortly to get booked in."


Prefer silence? Uncheck this and the AI simply stops replying.


Fallback message


This is a safety net for the rare case where something goes wrong — an error, or the AI can't determine whether to escalate. You shouldn't see it sent often.


Example: "Sorry, something went wrong on my end."


You can customize the fallback text to whatever you'd like.



Limit escalation to business hours


By default, human escalation can fire at any time. If you'd rather only hand off to your team when someone's around to respond, you can limit escalation to your business hours.


  • When enabled, new escalations only trigger during your company's configured Business Hours.
  • Outside those hours, the AI keeps responding as normal instead of handing off.
  • If you haven't enabled Business Hours, escalation stays available at all times.

This setting depends on Business Hours being enabled. If Business Hours aren't set up, the limit has no effect and escalation can fire around the clock.



Test before you go live


Once human escalation is turned on, use the built-in test module to try it in real time.


  1. Enter a message a lead might send — for example, "I want to talk to a manager so they can book me in now."
  2. Send it and watch what happens.
  3. Adjust the sensitivity or wording and run different scenarios to see how the AI responds.




If the criteria are met, the module confirms that human escalation was triggered and the message would be escalated. This lets you scenario-plan before turning it loose on real leads.



How your team gets notified


When human escalation fires, your team is notified in two places: the inbox and by email.


To make sure emails come through, confirm your notifications are on:


  1. Open Notifications, find your email or phone number, and confirm Human escalation required is enabled.

With notifications on, you'll get an email and / or SMS every time human escalation is required.


In the inbox, escalated conversations show a banner on the left sidebar:


  • Green banner — human escalation occurred and was resolved.
  • Red banner — human escalation occurred and is still open.




Resolving an escalation in the inbox


Open an escalated conversation to see the card, including any auto message the AI sent. From here you have a few options:


  • Mark it right or wrong — tell the AI whether the escalation was correct. This feedback helps improve future escalations.
  • Keep the AI off — leave the AI paused because you're going to reply manually.
  • Resume the AI — let the AI continue sending as scheduled.
  • Dismiss — exit out without taking action.



Once resolved, the card shows what happened and when — for example, that a team member left the AI off to respond manually, or that a lead requested to unsubscribe. This gives you a clear record of every escalation and how it was handled.



Need help? Contact us at support@leadtruffle.com or message the team via the chat on this article or in-app.

Updated on: 18/08/2026

Was this article helpful?

Share your feedback

Cancel

Thank you!