These people paid you and chose to stop. That makes them a different audience from cold signups — and most of them are not worth emailing at all.
A churn win-back sequence targets people who paid you and then cancelled. That is a different audience from signups who never converted, and it needs a different approach — these people know exactly what your product does and decided against continuing.
Fixing why they left returns more revenue than winning back the ones who already have.
A meaningful share of cancellations are failed or expired cards rather than decisions. Those people did not choose to leave, so a win-back email addressed to a decision they never made lands badly.
Card retry logic, an expiry warning and a dunning sequence recover this group at a far higher rate than any persuasion campaign, and they are a weekend of work. Check your failed payment rate first — the churn framework covers separating involuntary churn from the four voluntary types.
| Why they left | Win-back worth it? | Message |
|---|---|---|
| Never activated | No | They never saw it work. Nothing to return to. |
| The job ended | No | Structural. They finished. |
| Missing capability, now built | Yes | Name the specific thing that changed |
| Price at the time | Sometimes | Only if you now have a lower tier that fits |
| Switched to a competitor | Occasionally | Wait 6 months. Switching back has a cost. |
Two of five are usually not worth contacting at all, and sending to them dilutes your sender reputation for no return. If you do not know which bucket each cancellation falls into, that is the first job — getting churned customers to reply covers how to find out.
Message 1 — at cancellation. Not a win-back, a question. One line asking what happened, sent personally, with an explicit note that you are not trying to change their mind. This is where you learn which bucket they belong to, and it occasionally reverses a cancellation on its own.
Message 2 — at 30 days, only if something changed. The specific thing they mentioned now exists. No general update, no "we have been busy shipping". One sentence naming the change and one link.
Message 3 — at 90 days, optional. A short note if there has been a substantial change to the product or pricing since. If nothing meaningful has changed, do not send it.
That is the whole sequence. The restraint is deliberate — people who cancelled and then receive five emails do not come back, they mark you as spam.
A win-back discount is more defensible than a trial-expiry discount, because the customer already demonstrated they will pay and the alternative is zero revenue.
Two conditions. It should be time-bound rather than permanent, and it should not go to people who left over anything except price — a discount to someone who left because of a missing feature confirms you did not listen.
If price is a recurring cancellation reason among active users, the issue is more likely your value metric than your number. The value metric framework covers that, and the pricing playbook covers changing it.
Make the return path frictionless before sending anything. If a returning customer has to re-enter data, reconnect integrations or rebuild a workspace, the sequence cannot overcome that. Keeping data for 90 days after cancellation costs almost nothing and removes the largest barrier.
Win-back is a real but small number. Treat any recovery as a bonus rather than a plan, and judge the exercise on what you learn about why people left rather than on the recovery rate.
The comparison worth making: the same hours spent on activation reach a much larger group at a much higher conversion rate. Retention versus acquisition covers where effort pays best, and the churn guide covers the fixes for each type.
Three messages, not five. One at cancellation asking what happened, one at 30 days only if the specific thing they mentioned has changed, and an optional one at 90 days if there has been a substantial product or pricing change. Restraint matters more than volume.
Those who left over a missing capability you have since built, and sometimes those who left over price if you now have a tier that fits. Customers who never activated or whose underlying job ended are generally not worth contacting.
It is more defensible than a trial-expiry discount because the alternative is zero revenue. Keep it time-bound rather than permanent, and only send it to people who left over price — a discount to someone who left over a missing feature confirms you did not listen.
Win-back targets people who paid and then cancelled, so they know exactly what the product does. Cold signups mostly never reached first value, so there is nothing to return to. The two need different messages and different expectations.
Low single digits. Win-back is a small real number rather than a growth channel, and the same hours spent on activation reach a larger group at a higher conversion rate. Judge the exercise on what you learn about why people left.
Bring your cancellation reasons. Marcus sorts them into the buckets worth emailing and the ones to leave alone.
Try GhostCoach free →14-day free trial · cancel anytime · 30-day money-back on Lifetime