What is a win-back email?

A win-back email is an automated email, or short sequence of emails, sent to people who used to buy from you or engage with your brand and then stopped.

That definition is the settled part. The published guides disagree on the part that decides who actually enters the flow: what counts as gone quiet.

The win-back email, as published on 2026-09-02. One help page starts at 45 days after the last purchase. Another guide says 60-90 days without buying or engaging. A list-cleanup automation uses 90 days without opens, clicks or replies. One ecommerce automation waits 120, 240 and 360 days after the last purchase. A workflow template uses 180 days without purchase. Meanwhile, two sources say the right trigger is not a universal number at all: it is the store’s own repeat-purchase cycle.

What is a win-back email?

It is a lifecycle email sent after a customer or subscriber becomes inactive.

Mailchimp defines it broadly: the audience can be lapsed customers, subscribers or contacts. Klaviyo narrows the practical setup around a flow sent to customers who previously engaged with the brand and then have not interacted for a period of time.

So the short answer is:

The email can include a discount, but a discount is not the definition. The definition is the audience and the timing.

Is win-back the same as re-engagement?

Usually, yes.

Omnisend uses win-back and re-engagement together. Adobe also treats re-engagement and win-back as the same strategy. Mailchimp’s older automation label is “customer re-engagement” under a section for winning back lapsed customers.

There is a useful distinction, but the sources do not enforce it cleanly:

That distinction matters because a buyer who has not purchased is not the same person as a subscriber who has not opened. A customer can ignore email and still buy. A subscriber can open email and still never buy.

Is a win-back email only for previous buyers?

Not always.

Several purchase-triggered templates do work that way: someone bought, waited past the normal repeat-purchase window, and then entered a recovery sequence.

But the live pages do not reserve the phrase that tightly. The resource-page definition includes subscribers and contacts as well as customers. The deliverability version starts from email activity, not from order history. That means a win-back article, dashboard or report needs the audience written beside the term.

If the entry rule is no purchase, the flow is trying to recover revenue. If the entry rule is no open or click, the flow is trying to confirm whether the contact still belongs in normal sends.

What counts as inactive?

That is the disagreement.

Some sources define inactivity by purchase behavior. Drip calls the audience lapsed purchasers: people who bought before and have passed the typical repeat-purchase window. Its template segment uses at least one prior purchase and no purchase in the last 180 days.

Attentive starts even earlier for its triggered journey: no completed purchase in the past 45 days, then monthly follow-ups if the subscriber still has not bought.

Other sources define inactivity by engagement. ActiveCampaign builds its workflow around subscribed contacts who were sent campaigns in the last 90 days but did not open, click or reply. Adobe’s deliverability page gives the same 90-day rule of thumb for no open or click activity. Omnisend puts the trigger on customers who go quiet for more than 60-90 days and do not buy or engage with the brand’s content.

Those are different populations.

So “inactive customer” is not enough. The useful definition says inactive from what: buying, browsing, opening, clicking, replying, content engagement, or all of them.

Can a win-back email start after 45 days?

It can, if the source is using a purchase-risk model.

Klaviyo Academy describes a winback flow for customers who bought at some point and have not recently viewed products, added to cart, or purchased. The separate help-center source is the one that ties the first delay to the business’s average buying cycle.

The triggered-journey help page is the earliest fixed threshold in this source set: it applies to subscribers with no completed purchase in the past 45 days. That is not the same as saying every store should fire at 45 days. It says that one published template treats that point as win-back eligible.

Is 90 days the standard win-back trigger?

No. It is a common engagement rule of thumb, not a universal win-back rule. The published triggers in this source set run from 45 days without a purchase to 360 days, and the 90-day versions measure email engagement rather than orders.

The source set publishes all of these thresholds:

One reason those numbers disagree is straightforward: products do not have the same repurchase rhythm. A shampoo buyer can be lapsed while a sofa buyer is still behaving normally.

That is why Klaviyo’s help page gives the variable rule: find the average buying cycle for repeat customers, then set the first delay a little longer than that cycle. For a store with replenishable products, the right answer may be a replenishment flow before it is a win-back flow at all.

Why do some templates wait 180 days?

Because those templates are using purchase recency, not email engagement.

The workflow segment is built around at least one prior purchase and no purchase in the last 180 days. Mailchimp’s classic automation page goes longer: its customer re-engagement defaults wait 120, 240 and 360 days after the last purchase.

Those are not contradictions of the 90-day engagement rule. They are different clocks. A contact can be email-inactive at 90 days and still not be purchase-inactive for a product with a longer repeat cycle.

When is it a replenishment email instead?

When the product has a predictable use-up window.

If the customer is expected to need more before they are really gone, the email is closer to replenishment than win-back. That distinction matters for timing and tone. A replenishment reminder says the customer may be running low. A win-back says the customer has been away longer than expected.

The same customer can move through both states. A missed replenishment window can become a win-back trigger later, but the two emails are doing different jobs.

How many emails should a win-back sequence include?

Several sources use three emails, but that is not the only published answer.

The help-center pattern recommends three emails per recipient. The lifecycle guide describes a two-to-three email sequence over two to three weeks. The resource page shows the same shape: a check-in, a stronger offer, then a last-chance message.

But the templates do not all stop there. Drip’s workflow comes with five emails. Attentive’s setup instructions lay out four sends, while its FAQ says a typical win-back campaign has around five messages. ActiveCampaign’s version is not a neat per-recipient sequence at all: it batches contacts into a workflow and repeats branches as needed.

The practical rule is simpler than the source spread:

1. Start soft.
2. Increase the reason to return only if the person stays inactive.
3. End with a clean choice: buy, update preferences, unsubscribe, or stop receiving marketing.

The source set does not prove that a longer sequence is better. Some versions also treat non-response as a list-cleanup signal, not only a revenue recovery chance.

Should the first email include a discount?

Not by default.

The shared pattern starts softer than that. The first message can remind the customer what changed, show products related to the last order, ask for feedback, or give them a direct way to manage preferences. A discount is one possible reason to return.

The supported ladder starts softer, then increases the offer only for people who stay inactive.

What should the emails say?

The first message should explain why the person is hearing from you now.

The source set points to relevance signals: what the customer last bought, what they have missed, what changed, or what they may need next. It also points to a soft-to-stronger sequence rather than defining the flow by its discount.

A clean sequence looks like this:

The offer can get stronger, but it should not be the whole strategy. If the store has a real product cycle, timing and product relevance belong in the flow before the final offer.

Is a win-back email also a cleanup flow?

Often, yes.

A win-back flow is not only asking for the next purchase or click. It is also creating evidence about whether the person should keep receiving normal marketing.

If someone ignores the sequence, the next action should be different from business as usual: suppress, archive, unsubscribe, hold the subscription, or send a reconfirmation message.

What happens if they do not respond?

The next step is list hygiene.

The exact action differs by source. The list-cleanup automation unsubscribes contacts who do not engage with the win-back email. The triggered-journey help page says contacts who finish without engaging should be removed from email marketing unless they purchase or resubscribe on their own. The lifecycle guide says continued non-response after 60-90 days means preparing for sunsetting. The deliverability page gives a longer end state: no activity within the past year can put the subscription on hold before a reconfirmation ask.

The wording changes, but the shared rule is clear: do not keep mailing the same inactive group forever in the same way.

That matters because win-back is partly a deliverability tool. If a contact does not buy, open, click or reply after the recovery attempt, continuing normal sends can create sender-reputation or deliverability risk.

What should you measure?

Measure the outcome that matches the entry rule.

If the flow entered people because they stopped buying, the main result is purchase reactivation: how many bought again. If it entered people because they stopped engaging, the result is engagement: how many opened, clicked, replied, updated preferences or confirmed interest. If the flow is a cleanup step, the result is also the cleanup outcome: how many inactive contacts were suppressed, unsubscribed, removed, held, or asked to reconfirm.

This is where published win-back numbers get slippery. The deliverability page repeats an older Return Path finding: 12% open rate on re-engagement campaigns, 14% for normal campaigns, 24% reading the first re-engagement campaign, and around 45% reading later messages. That is an engagement statistic, not a purchase reactivation rate.

So a “win-back rate” without a denominator is not a useful number. Ask three questions before comparing it:

1. Was the audience inactive by purchase, engagement, or both?
2. Was success a purchase, an open/click/reply, or a confirmed opt-in?
3. How long after the first win-back email was success counted?

Without those answers, two stores can both report a win-back rate and be measuring different jobs.

What should the flow stop doing?

It should stop sending ordinary campaigns to people the recovery attempt already showed are not responding. That does not always mean a hard unsubscribe: the source set uses several end states — unsubscribe, remove from marketing, sunset, archive, hold, or reconfirm.

The common operating point is that unchanged sending is the wrong default after the win-back attempt fails.

The stop rule should be written before the flow launches. Otherwise a win-back sequence becomes one more set of emails layered on top of a list that was already not engaging.

When should you use a win-back email?

Use one when previous purchasers, subscribers, or contacts have passed the buying or engagement window you set, and you need to see whether they buy, re-engage, or move toward cleanup.

For ecommerce, write down:

The email itself is easy to name. The operating definition is the important part. A win-back flow is easier to judge when everyone can see who entered it, why they entered it, and what counts as won back.