Marketing Strategy & OperationsTemplate3 min readUpdated September 2026

Win-Back Emails for Lapsed Customers: A Three-Email Plan

A win-back campaign is a short email sequence sent to customers who've stopped buying, aimed at bringing them back or cleanly ending the relationship. Define "lapsed" by your normal purchase cycle, send two or three emails over about three weeks, and remove anyone who doesn't respond so you protect deliverability.

Win-back is often the cheapest revenue an email program can recover, because these people already know you. It also gets misused: blasting a discount at everyone inactive trains customers to wait for the coupon. The plan below avoids that and includes a way to measure whether the campaign caused any of the sales.

Vendors Covered in this Article

Disclosure: We may earn a commission if you buy through some links on this page. It doesn't change what we recommend.

How do you define a lapsed customer?

Use your own repurchase pattern, not a generic 90 days. Look at how long it usually takes a customer to buy a second and third time, and set the lapsed line at roughly one and a half to two times that gap.

For example, if most of your customers reorder every 30 days, someone silent for 60 days is lapsed. If you sell mattresses, a customer silent for 60 days is just a customer.

Split the lapsed group by past value. Someone who bought five times and stopped deserves a more personal email than someone who bought once at a steep discount. Keep at least two segments: high-value and everyone else.

The three-email sequence

Send these roughly a week apart:

  1. Email 1, the check-in: no offer. Mention what they bought, what's new since, and ask a direct question, such as whether something didn't work out. Replies here are useful research.
  2. Email 2, the reason to return: show what changed (new products, features, or service) and include a relevant recommendation based on their history.
  3. Email 3, the last call: a clear, honest final message. If you're going to make an offer, this is where it goes, with an expiry date that's real.

After email 3, stop. Anyone who hasn't opened or clicked any of the three moves to a suppression or sunset list.

Should you offer a discount in a win-back email?

Not by default. Try the first two emails without one. Many lapsed customers left for reasons a discount doesn't fix, such as a poor experience, a change in need, or forgetting about you.

If you do offer one, make it the last email, and test the size. Say you offer a fixed amount off to one half of the high-value segment and free shipping to the other half. Compare revenue per email, not redemption counts. Also check whether people who used the discount buy again afterward, or vanish again.

For subscription or software products, offering an extra month or a tailored onboarding call often works better than a price cut, and it doesn't reset your price expectations.

How to measure whether the campaign worked

Before you send, hold back a random ten to twenty percent of the lapsed list and send them nothing. After the campaign, compare purchases in the sent group against the holdout over the same period.

The gap between them is the revenue your emails caused. Without a holdout, you'll credit the campaign for purchases people would have made anyway.

Also track unsubscribes, spam complaints, and replies, and count a win only if the returning customers make a second purchase. A one-time coupon buyer who never returns isn't a win-back, it's a discounted sale.

When to stop emailing and clean the list

Emailing people who never engage hurts your sender reputation, and mailbox providers weigh engagement when deciding where your mail lands. After the sequence ends, suppress contacts who haven't opened or clicked in a long stretch, decided by your purchase cycle.

If the list is large, do it in stages: pause the least engaged segment first, watch deliverability, then continue. Consider a re-permission email for older contacts before you remove them. Keep purchase records separate from marketing consent, so a suppressed contact still receives order and account messages.

Executive Capability Standard

What Good Looks Like

Lapsed status is defined by your own repurchase cycle, the sequence is short and ends with suppression of non-responders, and a holdout group shows how much revenue the campaign caused.

Building The Capability (5-Stage Skill Ladder)

1. Learn:Calculate the typical days between first, second, and third orders so your lapsed definition rests on real behavior.
2. Do Manually:Export lapsed customers into two value segments, send the three emails from your email tool, and keep a random holdout.
3. Delegate:Give a lifecycle marketer ownership of the win-back program, including the holdout comparison and quarterly list cleaning.
4. Automate:Build a flow that enrolls customers automatically when they pass the lapsed threshold and exits them on purchase.
5. Buy:If you have a large catalog or many customers, use a lifecycle platform with predictive segments and dynamic offers, and verify holdout support in a demo.

How to Get Started

Disclosure: We may earn a commission if you buy through some links on this page. It doesn't change what we recommend.

Frequently Asked Questions

How long should a customer be inactive before a win-back email?

Base it on your typical repurchase gap, not a generic number. A common approach is one and a half to two times the usual time between orders. Products bought yearly need a much longer window than consumables.

How many win-back emails should I send?

Two or three, about a week apart. After the last one, suppress contacts who haven't engaged instead of continuing to mail them. Repeated mail to disengaged contacts can hurt deliverability for your whole list.

Does a win-back campaign need a discount?

No. Try a check-in and a useful reminder first. Add an offer only in the final email, and test it against a no-discount version. Discounts can teach customers to wait for coupons, so measure repeat purchases afterward.

How do I know a win-back campaign caused the sales?

Hold out a random slice of the lapsed list that receives no emails, then compare purchases between the groups over the same period. The difference is the lift your emails produced, rather than orders that would have happened anyway.

About the numbers

This guide doesn't quote a sourced benchmark. Figures in it are estimates or general guidance, so check them against your own numbers.

Related Guides