Win Back Lapsed Customers (And Know When to Stop)

A lapsed customer is not a lost customer. They already trusted you with a credit card once, which puts them miles ahead of a cold prospect who has never heard your name. But there is a window. Wait too long and the brand affinity goes cold, the email goes stale, and you end up shouting discounts at someone who moved on six months ago. The whole game is catching the drift early, making a real offer to come back, and being honest enough to walk away when they are gone.

First, define lapsed for YOUR store

There is no universal definition of a lapsed customer, and copying someone else's is the first mistake. Lapsed is a function of your purchase cycle - the natural rhythm of how often a happy customer buys from you again. A coffee subscription brand has a cycle measured in weeks. A mattress brand has one measured in years. Calling a coffee buyer lapsed at 90 days is smart. Calling a mattress buyer lapsed at 90 days is insane.

Find your real cycle before you build anything. In Klaviyo, look at the median time between first and second order for repeat customers. That median is your baseline. A reasonable rule of thumb: a customer is going quiet once they pass roughly 1.5x that median with no order, and they are properly lapsed somewhere around 2x to 3x. Those are starting points to set the timer, not laws. Adjust once you see how your own list behaves.

Spot the quiet ones before they're gone

Waiting for someone to fully lapse before you act is waiting too long. The early signal is not zero purchases - it is fading engagement while the clock runs. A customer who used to open every email and now opens none, who clicked through monthly and now ignores you, is drifting whether or not their last order date has crossed your threshold yet. Watch for the combination.

• Days since last order creeping past your going-quiet mark (the ~1.5x cycle number above).
• Email engagement falling off - they were opening and clicking, now they are silent across the last several sends.
• A skipped natural reorder - the consumable ran out, the season turned, the replenishment moment came and went with no purchase.
• No site activity - they have not browsed, even when you sent something they used to click.

Each of these on its own is noise. Stacked together, they are a customer telling you they are halfway out the door. That is the moment to act, not the moment three months later when the relationship is already dead.

What a winback sequence should actually offer

Here is the contrarian part. The default winback email is a discount, and a discount is usually the weakest thing you can lead with. Open with a code and you teach your best customers a bad lesson: ignore us long enough and a deal shows up. You also tell yourself a comforting lie - that the only reason they left was price. Usually it was not. They forgot, life got busy, a competitor caught their eye, or the product just did not stick. A discount answers a question nobody asked.

A good winback sequence is a short arc, not a single coupon. Three to four emails, spaced out, each doing a different job. Lead with the relationship and the reason, and hold the heaviest incentive in reserve for the people who need it. Here is a structure that works.

1. The check-in. No discount. Acknowledge the gap honestly ("It's been a while") and remind them what they liked - ideally the exact product they bought. Ask a real question or point them to what's new since they left. You are testing whether warmth alone brings them back.
2. The reason to return. Now give them news, not a code. A new collection, a restocked favorite, an improved version of the thing they owned, a use they may not have tried. Make coming back feel like catching up, not redeeming a coupon.
3. The incentive. If they're still quiet, now you make a real offer - and make it good enough to move someone who needed a push. A meaningful, clearly time-bound reason to act this week, not a permanent 10% that trains discount-waiting.
4. The last call. One honest final note. "We don't want to crowd your inbox - want to keep hearing from us?" This both recovers the people on the fence and cleanly sets up the sunset for the people who are gone.

Segment so the offer fits the customer

Not every lapsed customer deserves the same arc. The person who ordered five times and then went quiet is a different problem than the one-and-done buyer who never came back. Your highest-value lapsed customers have earned a warmer, more personal touch and probably a better offer - they were worth more, and they are worth fighting for. A single-purchase buyer can go into a lighter, more automated version of the same flow. Match the effort and the incentive to what the relationship was actually worth. Same skeleton, different weight.

When to sunset instead of emailing forever

This is the part most brands skip, and it quietly costs them. If someone has gone through your entire winback arc - check-in, reason, incentive, last call - and has not opened a single one, they are not a customer anymore. They are dead weight on your list. And dead weight is not harmless. Sending to people who never engage is exactly the signal Gmail, Yahoo, and Apple use to decide your whole list is unwanted. Keep mailing the unengaged and you drag your inbox placement down for the customers who DO want to hear from you.

So you sunset. A sunset is a deliberate rule that suppresses people who have stopped engaging entirely, so you stop mailing a void. A simple version: if a contact has not opened or clicked in a long stretch (often somewhere in the 90-to-180-day range, depending on your send frequency) and has been through the winback flow with zero response, you move them out of your regular sends. You are not deleting them. You are protecting your deliverability and your sender reputation by no longer paying the engagement penalty for someone who left.

Sunsetting feels like giving up. It is the opposite. A tight, engaged list of people who actually want your email lands in the inbox. A bloated list padded with ghosts lands in spam, taking your real customers down with it. Pruning the dead is how you protect the living.

Put it on rails

All of this should run as automated flows in Klaviyo, not as something you remember to do manually each quarter. A winback flow triggered off your going-quiet threshold, branched by customer value, with a sunset condition at the end - set it once and let it work. That is exactly the kind of flow Kaydence generates from your store: it reads your products and your brand voice, drafts the winback sequence and on-brand creative, and imports the flow straight into your own Klaviyo for you to review and push live. You approve every email and every send setting. The point is that the lapsed-customer arc - the part that protects both revenue and your inbox - actually gets built, instead of sitting on a someday list while good customers drift away for good.