You have already sent the "We miss you" email. It went out to everyone who had not bought in 90 days, it offered 15 percent off, and it pulled back a handful of customers who would probably have returned anyway. The rest stayed quiet. The problem was not the copy. The problem was that the campaign treated every lapsed customer as the same person with the same reason for leaving, fired on an arbitrary calendar, and led with the one thing that quietly damages your margin.

A Shopify win-back email campaign that actually reactivates lapsed customers is not a single broadcast. It is a sequence with logic behind every step, built on a definition of "lapsed" that matches how often people actually rebuy your product, and increasingly run across more than one channel. This guide covers the math, the segmentation, the four-message structure, and the mistakes that keep most win-back flows stuck below a 3 percent reactivation rate.

Why reactivating beats acquiring, in actual numbers

Start with the economics, because they decide how much you should be willing to invest in a win-back program.

Acquiring a new DTC customer means paying a customer acquisition cost (CAC) that, for most Shopify brands buying paid traffic, lands somewhere between 20 and 60 dollars before the customer has spent a cent. You are paying Meta or Google for a cold prospect, hoping they convert, and hoping the first order covers the ad spend. Many first orders do not.

Reactivating a lapsed customer is a different transaction entirely. They already know the brand, they have a stored profile, they have bought before, and they may still be on your owned channels. The marginal cost of a win-back email is close to zero, and even a multichannel sequence with an SMS or WhatsApp message costs cents per recipient. You are not buying attention. You are spending a fraction of one percent of CAC to re-engage someone who already trusted you enough to check out once.

The benchmark to anchor on: a well-built win-back program reactivates 3 to 10 percent of lapsed customers. The spread is wide because it depends on product quality, how recently the customer lapsed, and how well the list is segmented. A brand sending one generic discount blast to a two-year-old list will sit at the bottom. A brand running a logic-driven sequence against a freshly defined lapsed segment, across the right channels, reaches the top.

Run the math for your own store. If you have 5,000 lapsed customers with an average order value of 60 dollars, the difference between a 3 percent and an 8 percent reactivation rate is the difference between 150 and 400 recovered orders, or roughly 9,000 versus 24,000 dollars in revenue, from a sequence that costs almost nothing to send. That gap is why win-back deserves real attention rather than a recycled "come back" template.

Stop using "90 days." Define lapsed by your repurchase cycle

The single biggest reason win-back campaigns underperform is a lazy definition of "lapsed." Most stores inherit a generic 90-day rule from a blog post or an app default and never revisit it. That rule is wrong for almost every vertical because it ignores how often your customers actually rebuy.

Lapsed is not a fixed number of days. It is a deviation from a customer's expected behavior. The right way to define it is to look at your median time between orders for repeat customers, then flag someone as lapsed once they cross roughly 1.5 to 2 times that interval without ordering. Before they hit that threshold they are not lapsed, they are simply between purchases, and messaging them with a win-back offer is premature and trains discount-seeking behavior.

Practical starting windows by vertical:

• Consumables (supplements, coffee, pet food, skincare refills): 60 to 90 days. These products have a natural depletion cycle. If someone bought a 30-day supply and has not reordered by day 75, the gap is meaningful. The replenishment window is your signal.
• Fashion and apparel: 90 to 120 days. Purchase cadence is seasonal and irregular. A customer who bought in spring is not lapsed in July; they may simply be waiting for the next drop. Stretch the window so you do not pester active-but-quiet buyers.
• Beauty and cosmetics: 45 to 75 days. Many products run out faster than skincare or supplements, and the category has high competition, so a shorter window catches customers before a competitor's ad does.

These are starting points, not laws. The correct move is to pull your own data. In Shopify, the time-between-orders distribution for your repeat customers tells you the real cycle. If your supplement brand's median reorder is 38 days, your lapsed threshold might be 60 days, not 90. If your apparel customers average 140 days between orders, a 90-day win-back is firing at people who are perfectly normal.

Getting this window right is upstream of everything else. Send too early and you teach customers to wait for the coupon. Send too late and the relationship has already gone cold. For a deeper look at how behavioral timing beats calendar timing across the whole lifecycle, see our AI retention marketing playbook.

Segment the list with RFM before you write a word

Once you know who counts as lapsed, the next mistake is treating all of them as one audience. A customer who bought from you eight times and then went quiet is a completely different win-back problem than someone who bought once and never returned. Sending both the same message wastes your best opportunity and over-discounts your worst.

This is what RFM segmentation is for. RFM scores every customer on three axes: Recency (how long since the last order), Frequency (how many orders), and Monetary value (how much they have spent). Shopify's customer data and segment filters give you enough to build this without a separate data platform. The two lapsed segments that matter most:

• Slipping Champions. High frequency, high monetary value, but recency has crossed your lapsed threshold. These are your most valuable customers going quiet, and they almost never need a discount. They need acknowledgment, a reason to return, and possibly a check that nothing went wrong with a past order. Discounting a former VIP can actually insult them and signal that you do not recognize their history.
• One-time buyers who lapsed. Low frequency by definition, unknown loyalty, bought once and disappeared. This group is larger, colder, and more discount-responsive. A modest offer is appropriate here because you have no relationship to protect and the goal is purely to trigger a second purchase, which is the hardest and most valuable transition in DTC.

Splitting these two groups changes both the message and the economics. The Champions get a higher-touch, lower-discount sequence that protects margin and relationship. The one-time buyers get a more aggressive offer because the alternative is losing them entirely. If you collapse them into one campaign, you either under-serve the Champions or over-discount the masses. For more on how segmentation evolved from broad lists to behavioral cohorts, read the evolution of customer segmentation.

The four-message win-back sequence

Here is the sequence itself. Four messages, each with a distinct job, spaced over roughly two to three weeks. The logic matters more than the exact day counts, which you should tune to your repurchase cycle.

Message 1: Signal check, no offer

The first message is the one most brands get wrong by stuffing a discount into it. Do not. The job of message one is to acknowledge the gap and remind the customer of the product, without spending margin to do it.

Open by naming the absence honestly: it has been a while. Then re-anchor them to why they bought in the first place, the specific product or category they purchased, the benefit, a new arrival in that line, or social proof. Close with a low-friction question rather than a hard sell: are you still interested, or has something changed?

This message does real work even when it does not convert directly. A meaningful share of lapsed customers simply forgot about the brand, and a plain reminder reactivates them at full price. By holding the discount, you reserve it for the customers who genuinely need an incentive, and you avoid teaching your entire list that silence earns a coupon.

Message 2: Soft offer, sent only to those still cold

Roughly five to seven days later, message two goes only to the customers who did not respond to message one. This is the first time an offer appears, and it should be modest: 10 to 15 percent off, or equivalent value such as free shipping or a free sample with their next order.

Frame it as a welcome-back gesture, not a fire sale. The discount is a nudge to overcome the small inertia of returning, not the headline of the relationship. For your slipping Champions, consider replacing the discount entirely with something non-monetary: early access to a new product, a loyalty perk, or a genuine "we noticed you have not been around and wanted to check in." Margin preserved, relationship respected.

Message 3: FOMO plus a harder offer

Message three, around five to seven days after message two, raises the stakes for customers who are still cold. This is where scarcity and a stronger incentive belong, because the customer has now ignored both a reminder and a soft offer.

Two levers work here, often together: time-bound urgency (the offer expires in 48 hours) and a slightly larger incentive (a bigger discount or a bundle) for the most discount-responsive one-time buyers. Keep the bigger offer away from your Champions; they should never see your deepest discount, because protecting their lifetime value matters more than one reactivated order. The emotional driver is loss, the offer is about to disappear, not gain.

Message 4: The sunset message

The final message is counterintuitive and consistently the highest-performing surprise in the sequence: tell the customer you are removing them from your list.

Something close to "We have not heard from you in a while, so we will stop emailing you to keep your inbox clean. If you still want to hear from us, click here." This message works because it inverts the dynamic of every previous send. Instead of asking for attention, it threatens to withdraw, and loss aversion does the rest. Customers who ignored three promotional messages will click to stay subscribed, and a portion convert immediately.

The sunset message also does important hygiene work. The people who do not respond should genuinely be suppressed or moved to a low-frequency segment. Continuing to email customers who have ignored four messages drags down your deliverability and inflates spam complaints, which damages inbox placement for the active customers who do want to hear from you.

What not to do in a win-back campaign

The structure above already avoids the biggest errors, but they are worth naming directly because they are so common.

• Do not lead with a discount. A discount-first win-back trains customers to abandon on purpose. If going quiet reliably produces a 20 percent coupon, your savviest customers learn to stop buying until the coupon arrives. Watch your discount dependency rate, the share of reactivations that required a discount, and work to lower it over time.
• Do not send from a no-reply address. A win-back is a relationship message, and a no-reply@ sender signals the opposite. Use a real, monitored address. Some lapsed customers will reply with the actual reason they left, a defective order, a shipping problem, a price objection, and those replies are worth more than the campaign itself.
• Do not ignore channel preference. This is the quiet killer. A lapsed customer who stopped opening email has not necessarily stopped engaging with the brand; they may read every WhatsApp message and ignore their cluttered inbox. Firing all four messages into a channel the customer abandoned guarantees a low reactivation rate no matter how good the copy is.

Multichannel win-back: when to add SMS and WhatsApp

Email is the default win-back channel because it is cheap and information-rich, but it is rarely the only channel a lapsed customer is reachable on, and for some audiences it is the wrong primary channel entirely.

The decision rule is simple: follow the channel where the customer is still reachable, not the channel that is easiest for you to send from. In practice that means:

• Add SMS when a lapsed customer has a valid mobile number, has given consent, and email opens have dropped to zero. SMS open rates dwarf email, and a single well-timed text at the soft-offer or FOMO stage can reactivate customers who never saw your emails. Keep it short, respect quiet hours, and always include opt-out.
• Add or lead with WhatsApp for markets where it is the dominant channel. For DTC brands selling into MENAP, WhatsApp is frequently the primary way customers communicate, and a win-back sequence built only on email leaves most of the opportunity on the table. In those markets, WhatsApp should not be a supplement to the email sequence; it should be the spine of it, with email as the secondary touch.

The principle that ties this together is the same one behind defining lapsed correctly and segmenting with RFM: meet the customer where they actually are, not where your tooling defaults. A customer who opens WhatsApp but not email is not unreachable; they are reachable on a channel you have not used.

The metrics that actually tell you it is working

Opens and clicks are diagnostics, not outcomes. Three metrics tell you whether a win-back program is doing its job:

• Reactivation rate. The share of lapsed customers in the sequence who place an order within the campaign window. The 3 to 10 percent benchmark is your reference. Below 3 percent, look first at list age, channel match, and whether the sequence leaned too hard on discounts.
• Discount dependency rate. Of the customers you reactivated, what share required a discount to come back? A high number means you are buying revenue you might have earned at full price, and it is a warning that your sequence is training coupon-seeking. The goal is to reactivate as many as possible from messages one and four, the no-offer touches.
• Re-lapse rate at 90 days. A reactivation that lapses again within 90 days was a discount-driven blip, not a recovered customer. Tracking this stops you from celebrating a vanity spike. A genuinely won-back customer re-enters the normal lifecycle and starts buying on cycle again.

Together these three keep you honest. Reactivation rate tells you the sequence pulls people back, discount dependency tells you what it cost, and re-lapse rate tells you whether the win was real. For more on predicting which lapsed customers are worth the effort in the first place, see our guide to predictive analytics for customer retention.

Make it an operation, not a one-time build

The sequence in this guide works, but only if it runs continuously against a correctly defined, freshly segmented lapsed list, across the channel each customer actually opens. Doing that by hand means constantly recalculating repurchase windows, rebuilding RFM segments, watching for the moment each customer crosses the lapsed threshold, and firing the right message on the right channel before the relationship goes cold. Most lean DTC teams cannot keep that running every day, which is why win-back so often degrades into the occasional "We miss you" blast.

Tranthor's Reactivation Agent detects lapsed-customer signals automatically and runs this sequence the way it is meant to run: it watches each customer's purchase cycle, identifies the moment they cross from quiet to genuinely lapsed, segments slipping Champions from one-time buyers, and delivers the four-message sequence across email, SMS, and WhatsApp based on where the customer is actually reachable. Retention stops being a campaign you remember to build and becomes an operation that runs in the background. See how it fits your store on the Tranthor pricing page.