Skip to content
DIRI AI
All guides

Guide · Conversation practice

Telecom Cancellation Save Call Script: Find the Real Reason

A respectful cancellation call script: find the real reason, make one relevant offer, and make leaving easy if the answer is no.

When a customer calls to cancel a phone or internet subscription, they have usually already decided. Many agents respond by launching into a stack of offers, which makes the customer feel stonewalled. A good save call works differently. It treats cancellation as information, finds out what is behind it, and makes one relevant offer if there is one.

This article gives a structure and example wording. Adapt it to your own company's policies and to the rules that apply in your country on cancellations, contract terms and what you must tell a customer. Where those rules say you must process a request, process it.

The principle: respect first

A customer who calls to cancel is trying to do something simple. If you make it hard, even a good offer will feel like pressure. Two rules keep a save call honest.

  1. Never obstruct. If the customer insists on cancelling, help them do it.
  2. Offer only what is relevant. A discount for a customer whose problem is poor coverage is not a solution.

You can do both and still save a decent share of calls, because some customers are only waiting to be listened to.

Step 1: Acknowledge and confirm

Start by showing you understood, without arguing.

"I'm sorry to hear you're thinking of leaving. I can help with that. Before I do, may I ask what's behind the decision? It helps us improve."

Asking permission matters. If the customer says "no, just cancel", that is the answer.

Step 2: Find the real reason

The first reason given is often not the whole story. "It's too expensive" can mean the price rose, the customer is using less, or they received a competitor's offer. Use gentle, open questions.

  • "What made you decide to look at this now?"
  • "Was there something specific that changed?"
  • "How has the service been for you lately?"

Then listen. Take notes if you can, and repeat back what you heard: "So the signal in your flat has been dropping out since the spring, and that's what made you decide."

Step 3: Match one offer to the reason

Only now consider an offer. Match it to what the customer said, and make it once.

ReasonRelevant response
Price too highReview whether a cheaper plan or a plan change fits their usage
Not using the service muchSuggest a smaller plan that matches their usage
Coverage or speed problemsOffer to raise a technical check or explain what can be done
Moving houseExplain whether the service can move or how to end it cleanly
Found a competitorAsk what the competing offer includes, then compare honestly

Be transparent. If you suggest a cheaper plan, state the price, the length of any commitment and what changes. Do not hide terms, and do not present a trial rate as a permanent price.

"Based on what you told me, you use mostly Wi-Fi at home. A smaller data plan would bring your monthly cost down to [price], with no new contract period. Would you like me to explain what changes?"

Step 4: Accept the answer

If the customer says no, thank them and move on. A second and third offer after a clear "no" is where save calls turn into complaints.

"Understood, thank you for considering it. I'll go ahead with the cancellation. Here's what happens next."

Then explain the details in order: the end date, any final bill, returning equipment, and how they will receive confirmation. Clarity at this point leaves a better final impression than another pitch.

Step 5: Leave the door open

A customer who leaves on good terms may return. A short, genuine closing line is enough.

"If your situation changes, you are welcome to contact us again. Thank you for being a customer."

A full example

Customer: I want to cancel my mobile plan.

You: I'm sorry to hear that. I can help. May I ask what led to the decision?

Customer: The price went up, and I barely use the data.

You: That makes sense. How much data do you use in a normal month?

Customer: Maybe a couple of gigabytes.

You: In that case, a smaller plan would suit you better and would cost [price]. There's no new commitment period. Would you like me to go through it?

Customer: No, I've already got an offer elsewhere.

You: Understood, thank you for telling me. I'll set up the cancellation now. Let me explain what happens next.

The agent listened, made one relevant offer, and accepted the decision.

Mistakes to avoid

  • Reading a retention script regardless of reason. Customers can tell.
  • Making many offers in a row. One relevant offer is enough.
  • Arguing with the customer's choice. You are not there to prove them wrong.
  • Vague or hidden terms. If an offer has conditions, say them clearly.
  • Making cancelling hard. It damages trust and can breach your obligations.

Practise the hard moments

The difficult parts are staying calm when the customer is annoyed and not rushing to offer before you understand. Both are easy to rehearse. Role-play three customers: one who is angry about a price rise, one who is polite but decided, and one who is undecided and waiting to be convinced.

With DIRI AI an AI plays the customer. You speak by voice, and afterwards a report points to moments you could repeat, for example when you offered too early. Read how scoring works or listen to the sample dialogues. The Free plan gives you 3 calls a month of up to five minutes each, with no card.

Related: Telecom Subscription Sales Call Practice, How to Handle an Angry Caller, Refund Request Call Script.

FAQ

Should an agent always make a retention offer?

No. If the customer is clear and the reason is not something you can fix, accept the decision and help them leave smoothly. Offer only when it matches the reason given.

How many offers are appropriate in a cancellation call?

One relevant offer is usually enough. More than that after a clear "no" feels like pressure.

How can agents practise these calls?

Role-play with different customer moods, saying the awkward sentences aloud, ideally with an AI or human customer who gives feedback afterwards.