Rescheduling and Cancelation of Appointments

Rescheduling and Cancelation of Appointments

Booking an appointment is the easy part. Rescheduling or cancelling one is harder, because you have to find the existing appointment, verify the caller has the right to change it, and then make the right API call. This article covers what each calendar integration supports and the design patterns to handle reschedule and cancel safely.

What Each Integration Supports

These will be found in the 'Actions' Tab:
  • Generic Calendar / Generic Booking — book only. The agent finds the right calendar and texts the caller a booking link. It does not look up existing appointments, cannot reschedule, and cannot cancel.
  • Calendly (via MCP) — book, and with the MCP connector enabled, the agent can also look up and modify existing bookings. Reschedule and cancel require the caller's email (the account key on Calendly's side) and the event reference.
  • Zoho Bookings — Book, edit, and reschedule appointments through the Zoho MCP.

If you are on an older (non-MCP) Calendly setup, the agent v1 can only send a fresh booking link. To get live reschedule and cancel on the call, enable the MCP connector in your integration settings.

Why Verification Matters

A reschedule or cancellation changes state on the caller's behalf. You do not want an agent taking that action just because someone called in and said a name.

At minimum, verify one piece of information the real owner of the booking would know:
  • The email address on the booking (spell it back to confirm).
  • The phone number the booking was made from (can match against caller ID).
  • The original appointment date or time.

For high-stakes changes, stack two of these. The goal is not to turn the call into a security audit — it is to make it hard for someone to cancel a stranger's appointment on a whim.

A Good Reschedule Flow

A typical reschedule flow uses three states:

State 1 - Identify Appointment

Instruction: "Ask the caller for the email on their booking and the date of the appointment they want to change. Confirm the email by spelling it back."

Functions: Calendly search by email (MCP), proceed to verification.

State 2 - Confirm Intent

Instruction: "Read back the appointment you found and ask whether they want to reschedule or cancel. If reschedule, ask for the new preferred time."

Functions: Check availability (for reschedule), proceed to execute.

State 3 - Execute

Instruction: "Apply the change and confirm the new time or the cancellation to the caller. Text a confirmation."

Functions: Reschedule booking, cancel booking, send SMS confirmation, end call.

Info
Always confirm the email by spelling it back. "B as in boy, O, N, E" avoids the classic "Ben" vs "Ken" mistake that will cancel the wrong appointment. 

Safer Fallback: Transfer to a Human

If the caller cannot produce the email on file or the agent cannot find a matching appointment, do not let the agent keep guessing. Transition to a Transfer state and hand off to a human. For most businesses, the cost of one bad cancellation is higher than the cost of a transferred call.


Tips

  • Put all reschedule/cancel tools in the same state once the caller is verified. The agent decides the order based on what the caller asked for.
  • Always text a confirmation after the change. Voice confirmations are easy to mishear.
  • Log the call. Reschedule and cancel actions are exactly the kind of thing you want on tape if a customer disputes it later.

Still have questions? Submit a ticket and we will help you design a safe reschedule flow.

    • Related Articles

    • OneCloud AI Agent | Generic Calendar Configuration

      OneCloud AI Agent | Generic Calendar Configuration Give your AI-Agent the ability to easily share a single booking link with Generic Calendar. This integration makes it simple for customers to schedule appointments with you directly, providing a ...

    • Transition Descriptions as Tools

      Transitions are the arrows between states in your flow. To the AI Agent, each transition looks like a tool it can call. The description you write on a transition is what the agent reads to decide whether to fire it. A good description is the ...

    • State based vs Prompt Engineering

      There are two ways to control how your AI Agent behaves on a call: 1. Write one long system prompt that tries to cover every situation or 2. Build a flow with states that change the agent's focus as the conversation moves forward. This article ...

    • How to Setup AI Agent

      Welcome to your AI Agent setup journey! This guide will walk you through creating a virtual agent that represents your business professionally and handles calls intelligently. Getting Started in 4 Easy Steps Configure Your AI Agent Customize ...

    • Building Conversation Flows

      A flow is the conversation blueprint for your AI Agent. It defines the stages of a phone call, what the agent does at each stage, which tools it can use, and when the conversation moves forward. Think of it like a flowchart: each box is a stage the ...