A win-back campaign goes after customers who have ended the relationship. Some canceled a subscription or closed an account. Others let a paid membership lapse or moved their service to a rival. That sets it apart from a reactivation campaign, which targets members who are still enrolled but have stopped buying. The win-back customer has made a decision. The campaign has to give them a reason to reverse it.
How is a win-back campaign different from a reactivation campaign?
The two get mixed up because both target people who stopped buying. The difference is whether the customer is still in the program. That one fact changes almost every part of the campaign, as the table shows.
| Win-back campaign | Reactivation campaign | |
|---|---|---|
| Who it targets | Customers who canceled, closed an account or did not renew | Enrolled members with no recent activity |
| What starts it | An exit event with a date, such as a cancellation | A gap longer than the member's normal buying cycle |
| What the customer still holds | Often nothing, where the terms void points and status on exit | A live account, a points balance and a tier |
| Main lever | A reason to come back that answers why they left | A reminder of the balance and the next reward |
| Success measure | Returns above a holdout, and how long returners stay | Members who resume activity above a holdout |
When should a win-back campaign start?
Start from the exit event. A cancellation or a non-renewal gives an exact date and often a stated reason. Record that reason at exit, because it shapes the offer. Kumar, Bhagwat and Zhang studied eight years of data from a US telecom firm. They found that a lost customer's first-lifetime behavior, their reason for leaving and the type of offer all bear on whether they return. The stronger the first relationship, the more likely the customer is to accept a win-back offer.
Split the leavers by type before any offer goes out. A customer whose card failed did not choose to leave. That is a billing fix, and a discount there wastes margin. A customer who canceled on purpose needs a reason to return. Members who opted out of marketing, or who left over a complaint, belong in their own groups. Then run a short sequence in the channels each group still accepts. Stop when a wave no longer beats its holdout.
What makes a good win-back offer?
The offer should answer the reason the customer left. Price is one lever, and a fix for the problem that drove them out is another. The Kumar study found that an offer pairing a discount with a service upgrade was the most effective at winning customers back. Customers who left over price were worth winning back too. They can be less profitable than customers who left over service, but they stay longer once they return.
Price the return in two steps. Thomas, Blattberg and Fox modeled both the return and the length of the second tenure. In their application, the best pricing was a low price to win the customer back and higher prices once they had returned.
Some of the strongest levers cost little. Netflix tells members who restart that their profiles, My List, game saves and payment information will still be there. Keeping a returner's history makes coming back easy. Some loyalty terms work the other way. Marriott Bonvoy terms say a member who cancels can reapply, but forfeited points and elite status are not reinstated. A program that wants leavers back can choose to restore status or points as the offer.
Set rules so current customers cannot game the deal. Disney+ offers its six-month bundle deal to new and eligible returning subscribers. A plan that is set to cancel does not qualify. That keeps a current customer from canceling just to get the lower price.
How much should you spend to win a customer back?
Cap the offer at what a returning customer is worth, adjusted for those who would have come back anyway. Some leavers return on their own, and each one who takes the offer is cost with no gain. The cap per returner is the margin a returner earns in their second tenure, times the share of returns the campaign caused.
The pool is large enough to plan for. Recurly, a subscription billing platform, reports that 1 in 4 new sign-ups are returning subscribers. Its report draws on data from 76 million subscribers. Treat returners as their own acquisition channel, with a budget and a cost per return. The retention economics guide covers the margin side.
How do you measure a win-back campaign?
Measure against a holdout. Pick a random slice of the eligible leavers and send them nothing. Compare return rates over the same window. The gap is what the campaign caused. Control group testing explains the method.
For illustration, a streaming service has 40,000 customers who canceled last quarter. It holds out 4,000 and offers the other 36,000 three months at $15 instead of $30. Within 90 days, 9% of the offer group returns, against 5% of the holdout. That is 3,240 returns, of which 1,440 are extra returns the offer caused.
Every returner takes the $45 discount, so the offer costs $145,800. Spread across all returns, that looks like $45 each. Spread across the extra returns, it is $101.25 each. Say a returner earns $120 in margin over their second tenure. The break-even discount is $120 times the share of returns the offer caused (1,440 of 3,240). That comes to about $53. The $45 offer clears it with little room.
Then track how long returners stay and what they spend. A cheap offer that brings back customers who leave again within a month is a loss.
What rules apply to win-back offers for canceled subscriptions?
Subscription law covers both the save offer at cancellation and the rejoin that follows. In California, Business and Professions Code 17602 lets a business show a discount or other offer when someone cancels online. The business has to display a prominent "click to cancel" link or button at the same time. By phone, a business that makes an offer must first tell the caller they can finish canceling at any time by saying so.
The same section covers discounted rejoin prices. If a promotional price lasts more than 31 days, the business must send a notice 3 to 21 days before it ends. The section's 2024 amendments apply to contracts made, amended or extended on or after July 1, 2025.
Federal law sets the floor for the rejoin itself. ROSCA requires a business that sells a recurring plan online to disclose all material terms before taking billing details. It also requires express informed consent before the charge and a simple way to stop the charges. Plan the restart flow with those duties in mind, even when the card is already on file. The FTC settlement with Amazon in September 2025 requires a way to cancel Prime by the same method customers used to sign up.
Contact rules limit who you can reach. Under CAN-SPAM, a business must honor an email opt-out within 10 business days. Members and subscribers keep that right. In the UK, the ICO soft opt-in lets you email previous customers about similar products if they bought recently and were offered an opt-out. Rules differ by country, so check each market before a campaign goes out.
How GRAVTY handles win-back campaigns
GRAVTY flags disengagement and churn risk early enough to act on it. Member 360 resolves one identity per member across email, phone, card and app. That gives a win-back team one record of what each leaver did before they left.
Offers and their rules are set up in Visual Rules, the patented no-code rules engine. Each rule is simulated before it goes live, and every version has an audit trail. Agentic AI Compass runs offer tests and simulates outcomes first. A team can check what an offer will cost before it reaches leavers. GRAVTY runs 500M+ members in production.
Frequently asked questions
What is an example of a win-back campaign?
One example is Disney+, which offers a cheaper six-month bundle to returning subscribers who qualify. The offer page excludes plans that are set to cancel, so current customers cannot cancel to claim it. Netflix shows a version with no discount. Members who restart find their profiles and My List still there.
How soon after a customer cancels should you send a win-back offer?
Send the first message while the relationship is recent, then test the timing of later waves against a holdout. In the UK, the ICO ties the soft opt-in for email to customers who bought recently. That is one more reason to start early.
What is a good win-back rate?
A good win-back rate is one that beats the holdout group by more than the offer costs. A raw return rate counts customers who would have come back anyway. Compare the gap in returns with the margin those extra returners earn.
Can you send win-back emails to customers who unsubscribed?
No, an email opt-out still stands after the customer leaves. The FTC's CAN-SPAM guide says opt-outs must be honored within 10 business days. Being a member or subscriber does not remove that right. Use other channels only where that customer's consent still holds.
Should a win-back offer restore lost points or status?
Restore them when the margin you expect from the returner covers the cost. Some programs void both on exit. Marriott Bonvoy terms say forfeited points and elite status are not reinstated when a member reapplies. A program with that rule can still offer restored status as the reason to return.
What is the difference between a save offer and a win-back offer?
A save offer comes while the customer is canceling, and a win-back offer comes once they have left. In California, the law says an online save offer must sit next to a prominent "click to cancel" link or button.