Use cases
Policy cancellations and auto-renewals
Handle policy cancellations and auto-renewals automatically, from retention to processing.
Tags
AI Agent
Inbound
Channel
Chat
Voice
Category
Quotes and policies
Account handling
Industry
Insurance
What it does
Customers who want to cancel, or who are surprised by an auto-renewal, are often open to staying, but only if the conversation happens quickly. Handled slowly or poorly, it's a churn event.
The Gradient Labs AI agent picks up cancellation and renewal contacts, checks the customer's policy status and cover type, and works through the options. If it's within the cooling-off period, the agent explains the refund (minus time on cover and the setup fee).
Outside it, the agent explains how charges work based on their policy. Before processing anything, the agent asks why the customer wants to cancel and tries to address the concern.
For auto-renewal, it reads the current status and expiry date, then walks the customer through changing it. Pay-as-you-go customers get guided through in-app cancellation. Customer intent is handled the moment it's expressed, not when an agent becomes available.
How it works
The Gradient Labs AI agent pulls live policy data from customer resources: status, product type, platform, expiry date, and auto-renewal status. It uses these to determine the right path for each customer.
When a cancellation needs to be processed by your team, the agent triggers Ask a Human, a tool call that hands the conversation to your live chat team with a full summary attached. The customer doesn't repeat themselves, and your team picks up with all the context they need.
The agent checks whether your team is online before handing off. If they're offline, it tells the customer when to expect a response. For auto-renewal changes, the agent calculates whether the customer is within 72 hours of expiry. If they are, it hands off to a human automatically. If not, it either creates a case for your team to action or walks the customer through toggling the setting themselves.
See it in action
Explore other use cases