Win-Back Email Strategy: How to Bring Lapsed Customers Back
The most important part of a win-back email is not the email.
It is deciding who should receive it.
A customer who bought from you eight times, used to order every month and disappeared six weeks ago should not receive the same sequence as someone who bought once on Black Friday two years ago.
Yet most win-back flows begin with one rule:
Has not purchased in 90 days.
Then everybody gets the same sequence.
That is convenient.
It is also where most of the strategic value disappears.
A strong win-back strategy should answer five questions before anyone writes copy:
- Is this customer genuinely lapsed?
- Is the relationship worth recovering?
- Why would they reasonably return now?
- What should we test before discounting?
- How will we know whether the campaign actually changed behaviour?
Only then should the campaign be built.
Start with the buying cycle
Recent win-back guidance increasingly agrees on one principle: timing should follow how customers actually buy.
Klaviyo recommends setting win-back delays just beyond the normal buying cycle, while newer ecommerce guidance such as Affinsy’s 2026 analysis argues for product-level lapse thresholds using median reorder gaps rather than one universal date. (affinsy.com)
That distinction matters.
A coffee customer who buys every 28 days is probably meaningfully late at day 60.
A customer who bought a premium cookware set may not be.
Same inactivity.
Different interpretation.
Separate the major win-back cohorts
At minimum, I would split:
Recent repeat lapse
Used to buy repeatedly.
Now behaviour broke.
One-time second-purchase overdue
Never developed a repeat relationship.
Former VIP
Historically valuable.
Needs more thoughtful treatment.
Replenishment overdue
Should reasonably need more product.
Cancelled subscriber
Has an explicit cancellation event and possibly a cancellation reason.
Deep lapse
Long absence, weaker brand memory, higher contact risk.
These groups do not need the same sequence.
Build the Return Reason before the copy
The campaign should answer:
Why should this customer come back now?
Possible Return Reasons include:
- replenishment,
- a genuinely relevant new product,
- convenience,
- education,
- recognition,
- changed product availability,
- a better subscription option,
- an incentive.
Notice that “we miss you” is not on that list.
It may be acceptable language.
It is not usually the reason the customer buys.
A practical sequence architecture
Most ecommerce brands do not need seven win-back emails.
A focused sequence is usually stronger.
Klaviyo’s historical research found that many of its strongest revenue-per-recipient win-back programs used three emails or fewer, while current re-engagement guidance still commonly recommends short multi-message sequences rather than endless follow-up. (marketingdive.com)
A useful structure is:
Email 1 — Relevance
Remind the customer why the brand is worth noticing again.
No discount by default.
Email 2 — Value or change
Show:
- a product update,
- relevant recommendation,
- new use case,
- proof.
Email 3 — Stronger return reason
If an incentive is justified, this is where it can become more useful.
Final touch — Decision
For colder customers:
- final offer,
- preference check,
- reduced-frequency option,
- suppression.
Not every cohort needs four emails.
The sequence should be as long as the commercial logic requires.
Why I would not discount first
Suppose a customer is slightly beyond their normal reorder window.
You lead with 20% off.
They buy.
Did the discount create the purchase?
Maybe.
Or maybe the customer needed only a reminder.
Starting with value gives you a chance to recover customers without paying for behaviour that might have happened anyway.
This is why many current win-back practitioners recommend escalating incentives rather than leading with the largest one. (brianroseman.com)
Measure purchases, not opens
An open does not mean the customer is back.
Track:
- reactivation rate,
- revenue per recipient,
- incremental lift,
- average order value,
- contribution margin,
- unsubscribe rate,
- complaints.
Where possible, use a holdout group.
If the treatment group buys at 6% and the holdout buys at 3%, the interesting result is not the full 6%.
It is the estimated 3 percentage-point lift.
Protect sender reputation
Win-back audiences naturally contain more disengaged customers.
That means sending too broadly can damage deliverability.
Google and Yahoo both require stronger authentication and low spam rates for bulk senders, and both make clear that unwanted mail can damage reputation. (marketingdive.com)
Start with the strongest cohort.
Expand only if the result and send health justify it.
Case example: Tibi
Klaviyo reports that Tibi moved from a generic win-back flow toward recency, frequency and monetary-value triggered journeys for valuable customers moving into weaker states.
Its top refreshed win-back flows reportedly generated more than twice the revenue of the prior flow in their first full month. Vendor-reported, but strategically useful. (klaviyo.com)
The lesson is not:
Use Klaviyo.
It is:
Better customer-state logic can outperform one generic inactivity trigger.
Frequently asked questions
What is a win-back email?
A win-back email is a message sent to a customer who has stopped buying or engaging, with the goal of bringing them back into active purchasing behaviour.
How many win-back emails should I send?
Usually two to four is enough for most ecommerce scenarios. The right number depends on customer value, lapse depth and whether the sequence includes a final sunset or preference message.
When should a win-back email be sent?
After the customer has moved beyond their expected buying cycle—not simply after a universal 60- or 90-day delay.
Should the first win-back email include a discount?
Not automatically. Start with relevance and value unless the economics clearly justify an immediate incentive.
What is a good win-back rate?
There is no universal rate. Category, customer history, lapse depth and incentive all materially change the result.
The principle
A win-back email is not the strategy.
It is the message that comes after the strategy.
Get the customer wrong and even great copy will struggle.
Get the customer, timing and Return Reason right and the email becomes much easier to write.
Related reading
Why “We Miss You” Emails Often Fail
Should You Discount Lapsed Customers?