Winback Email Campaigns: The Klaviyo Playbook for Lapsed Buyers
Somewhere in your Klaviyo account sits a segment worth more than your entire prospect list: customers who bought from you, liked the product enough not to return it, and then went quiet. They already trust you, they already know shipping works, and reaching them costs nothing. Most brands send them the same campaigns as everyone else, watch them not open, and eventually let the sunset flow suppress them without ever making a real attempt.
A winback campaign is that real attempt: a purpose-built sequence that treats lapsed customers as the distinct, high-value audience they are. Done in October, it has a second payoff, because every customer you reactivate now is engaged and mailable for Black Friday instead of sitting in your suppressed segment while your biggest offers of the year go out.
Winback is not your sunset flow
The two get conflated constantly, and the confusion costs revenue. The sunset flow is a deliverability tool: it targets subscribers disengaged with your email regardless of purchase history, makes a last attempt, and suppresses non-responders to protect your sender reputation.
Winback targets lapsed buyers. The entry condition is about purchase behavior, not email engagement: someone whose last order has passed your typical repurchase window. Many of them still open your emails. The sunset flow asks "should I stop mailing you?" The winback asks "what would bring you back?" Different question, different audience, different economics.
Define lapsed from your data, not a default
The biggest winback mistake is borrowing someone else's timeline. A coffee brand's customer is lapsed at 60 days; a furniture brand's customer isn't lapsed at 300. Pull your actual repurchase curve: the median days between first and second orders, and between subsequent orders. Your winback trigger belongs at roughly 1.5x to 2x that median.
If you have enough order history, Klaviyo's predictive fields make this sharper. Expected next order date and churn risk let you trigger the winback when this customer is overdue rather than when a calendar average says so; we covered how those models work in our guide to Klaviyo predictive analytics.
The three-send winback arc
Winbacks work as a short escalation, not a drip. Three sends over two to three weeks:
- Email 1: the reminder, no discount. Lead with what's changed since they left: new products, restocks, improvements, best sellers they haven't seen. A meaningful share of lapsed customers just drifted, and news is enough to bring them back at full margin.
- Email 2: the incentive. Five to seven days later, for non-purchasers only. Make the offer concrete and personal: a dollar amount off, framed around returning ("$15 to come see what's new"). Dollar-off typically beats percentage-off here because the value is instantly legible.
- Email 3: the deadline. The same offer with an expiration and a plain-spoken close. Honest urgency, then stop. Customers who don't respond to three purpose-built sends flow onward to the sunset process rather than getting a fourth and fifth nag.
Segment the arc by customer value. A one-time buyer from eleven months ago and a former VIP with eight orders deserve different treatment: the VIP gets a warmer first email, a stronger offer, and ideally a founder's voice. Your segmentation structure should already separate these tiers; the winback is where that structure pays.
Flow or campaign? Run both
The evergreen version runs as a Klaviyo flow, triggered as each customer crosses your lapsed threshold, so reactivation happens continuously without anyone remembering to send it. The campaign version is the seasonal sweep: once or twice a year, a one-off push to the whole lapsed segment. October is the obvious moment, with BFCM as the built-in story: "You'll want to be on this list next month." Reactivated customers then see your entire holiday calendar as engaged subscribers instead of suppressed ones. If you're building the whole pre-season sequence, our free BFCM playbook covers what surrounds the winback: segments, flows, and the send calendar.
Measure it like a program, not an email
- Reactivation rate: The share of entrants who purchase during the arc. Healthy programs recover 3% to 8% of lapsing customers, the same benchmark we use in our retention marketing guide, which sounds small until you price the segment.
- Revenue per entrant: The honest comparison metric against other flows, and the number that justifies the incentive spend.
- Post-winback behavior: Watch whether reactivated customers keep buying or only surface for discounts. If it's discounts only, your email 1 needs to work harder before the offer appears.
To see what the lapsed segment is actually worth before you build, run your list and order numbers through the email revenue potential calculator; unworked winback revenue is usually one of the biggest gaps it surfaces.
FAQ
When should a customer count as lapsed?
At roughly 1.5x to 2x your median repurchase interval, pulled from your own order data. For many DTC brands that lands between 90 and 180 days, but consumables run shorter and considered purchases much longer.
Should the first winback email include a discount?
No. Lead with novelty and let the customers who only needed a reminder come back at full margin. The incentive goes in email 2, targeted only at those who didn't respond.
Is October too late to run a winback before BFCM?
It's the ideal window. A three-send arc launched in early October completes weeks before Black Friday, and everyone it reactivates receives your holiday calendar as an engaged subscriber.
What happens to customers the winback doesn't recover?
They continue into your sunset process, which makes a final engagement-based attempt and then suppresses them to protect deliverability. Winback and sunset work as a sequence, not as substitutes.
Your cheapest revenue is sitting in a segment
Acquisition costs keep climbing; the customers you already won don't. Define lapsed from your data, run the three-send arc, keep the evergreen flow on afterward, and go into BFCM with a bigger engaged list than you had in September.
Want us to size your lapsed segment and build the arc with you? Book a 30-minute call and we'll pull the numbers together.