Win-Back Campaigns: How to Bring Back Lapsed Customers Without Discounting Everything

A good win-back email campaign brings lapsed customers back by reaching them at the right moment with something relevant, and saves discounts for the customers and moments where they are actually needed. The recipe is simple: define "lapsed" using your own repurchase data, segment by customer value, lead with new products and useful content, and use a holdout group to prove the flow is creating orders rather than just taking credit for them.

Most brands do the opposite. They send everyone who has not bought in 90 days a 20% code, which trains customers to wait and quietly erodes margin.

In short

• Set your lapse point from your median repurchase gap, not a round number.
• Treat one-time buyers and loyal customers differently.
• Use non-discount levers first. Discount last, and selectively.
• Measure against a holdout group, then suppress people who never respond.

Step 1: Define "lapsed" for your store

A skincare brand selling a 30-day product and a furniture brand have very different buying cycles, so a single 90-day rule rarely fits. Instead:

1. Export orders from Shopify for the last 12 to 24 months.
2. For customers with at least two orders, calculate the days between their first and second order.
3. Find the median. This is your typical repurchase gap.
4. Set your win-back trigger at roughly 1.5 times that gap, and a "lost" threshold at around 3 times.

For example, if a hypothetical brand's median gap is 40 days, the win-back flow starts around day 60 and customers past about day 120 are treated as lost. Shopify's RFM customer analysis report gives a quick sense check. Its groups include "At risk", "Almost lost", "Previously loyal" and "Dormant".

If your Klaviyo account qualifies for predictive analytics (Klaviyo requires at least 500 customers with orders, 180 days of order history and recent orders), you can trigger on each customer's expected date of next order instead of a store-wide number.

Step 2: Segment lapsed customers by value

This split matters because the maths is different. Losing a VIP costs far more than losing a one-time buyer, so a richer offer can be justified. For one-time buyers, the aim is often simply to get the second order, which is where retention usually starts.

Step 3: Lead with levers that are not discounts

Before offering money off, try the things that remind people why they bought in the first place:

• What is new: launches, new shades or sizes, improved formulas.
• Restock and replenishment: "The serum you bought in March is back" or "You are probably running low".
• Education: how to get more from the product they already own, routines, styling guides.
• Social proof: recent reviews or customer photos for the category they bought.
• Service perks: free shipping on the next order, a free sample, or prepaid perks in place of COD fees.
• Loyalty points: "You have points waiting" often works better than a new code.
• Early access: first look at a sale or launch for past customers.

These protect margin, and they also give you real information about why the customer drifted.

Step 4: Build the sequence

Filter every step on "placed order since starting this flow" so buyers exit immediately and move into your post-purchase flow. In the final email, a one-question survey ("Why haven't you ordered recently?") often turns up fixable issues such as sizing, delivery or price.

Example angles for each email

Here is how the sequence might read for a hypothetical haircare brand whose customer last bought a shampoo and conditioner set:

• Email 1, subject "Your hair routine, updated": introduces a new leave-in serum that pairs with what they bought, and links to their past products for a one-click reorder.
• Email 2, subject "What 400 customers said about the serum": short reviews, a before-and-after routine and a three-step usage guide. If you do not have that many reviews, use the real number.
• Email 3, subject "A thank-you for coming back": free shipping and a free travel-size sample on the next order, rather than a percentage off.
• Email 4, subject "Should we stay in touch?": reminds them the offer ends soon, asks one feedback question and offers to send fewer emails instead of none.

Notice that the first two emails sell nothing new at a lower price. They sell relevance. Only the third introduces an incentive, and it is one that costs less than a blanket discount.

Where WhatsApp fits

For opted-in VIPs, one WhatsApp message can outperform another email. Meta classes this as a marketing message, priced at ₹0.8631 per delivered message to Indian numbers on its rate card effective 1 July 2026, before GST and provider fees. Only message customers who have opted in, as required by the WhatsApp Business Messaging Policy, and keep it to one touch.

Step 5: Measure incrementality

Attributed revenue flatters win-back flows, because some customers would have come back anyway. The honest test is a holdout:

1. Randomly exclude around 10% of eligible customers from the flow. In Klaviyo, one way is to assign each profile a random group property and use a conditional split on it.
2. After 60 to 90 days, compare reactivation rate and revenue per customer between the two groups.
3. Subtract incentive costs from the difference to see true profit impact.

A worked example: suppose 5,000 lapsed customers enter the flow and 500 are held out. If 8% of the mailed group reorders against 5% of the holdout, the flow created about 3 extra orders per 100 customers, around 135 incremental orders from 4,500 people. If most of those used a 20% code, compare that discount cost with the margin on 135 orders before deciding the discount stays. These numbers are illustrative. Your own test will tell you.

Step 6: Suppress the ones who do not come back

Customers who ignore the whole sequence and have stopped opening or clicking should move into a sunset flow, then be suppressed if they still do not engage. This protects inbox placement. In Klaviyo it also lowers your bill, because suppressed profiles stop counting towards billable active profiles once processed.

Common win-back mistakes

• Using one lapse window for every product and customer type.
• Leading with the biggest discount you can afford.
• Sending win-back emails during a sitewide sale, when everyone gets a discount anyway.
• Never checking whether win-back customers stick around for a third order.

For subscription brands, the same thinking applies to cancelled subscribers. Our guide to increasing subscription LTV covers that side.

Winning customers back profitably

The best win-back programmes feel personal and relevant, and they discount only where the numbers justify it. If you want help building segments, sequences and holdout tests in Klaviyo, see how we approach Klaviyo retention or explore our email marketing services.