- Winback email #3 earns about 2x winback email #1 ($0.23 vs $0.11 per person) in pooled data from 4 consumable brands.
- Cart abandonment works the same way. Email #5 still earns about $0.25 per person, the same as email #3.
- Welcome is the opposite. Email #1 earns about 30x emails #2 to #5. Keep welcome short, keep cart and winback long.
- For consumables, I'd build winback as four or five emails, with the incentive saved for the middle of the flow, not the first email.
How long should a winback flow be, based on the data?
When I pooled the data from 4 consumable brands, winback was one of the flows where the later emails did more of the work, not less.
- Winback email #1: about $0.11 per person.
- Winback email #3: about $0.23 per person.
That's about 2x. If you cut your winback at two emails, you never send the one that earns the most.
The same thing shows up in cart abandonment. Email #5 still earned about $0.25 per person, the same as email #3. People keep buying deep into both flows.
Why do later winback emails earn more?
My read on it: a lapsed customer isn't a cold lead. They already know you, they've used the product, and something pulled them away. Maybe they ran out and forgot, maybe they tried a competitor, maybe life got busy.
The first email is a nudge, and a lot of people aren't ready yet. By the third, it's been a while, they've seen your name a few times, and there's often a better reason to come back on the table. That's when they move.
Whatever the reason, the numbers say the same thing. The back half of the flow is where the money is.
Winback vs cart vs welcome: which flows should be long?
Not every flow works this way. Welcome is the exact opposite.
| Flow | Early email | Later email | What to do |
|---|---|---|---|
| Welcome | #1: about $4.50 per person | #2 to #5: about $0.10 each | Keep it short. Put the effort into email #1. |
| Cart abandonment | #3: about $0.25 per person | #5: about $0.25 per person | Keep it long. |
| Winback | #1: about $0.11 per person | #3: about $0.23 per person | Keep it long. |
So my rule is simple. Keep welcome short. Keep cart and winback long.
The difference is where the person is. A new subscriber is paying the most attention right after they sign up, then it fades. A lapsed customer or a cart abandoner already wanted the product. They just need more time and more reasons.
A winback flow structure for consumable brands
Here's how I'd set it up for a brand selling something people use up. This is a starting point, not a rule. Adjust the timing to how fast your product actually runs out.
When it starts. Winback should kick in once a customer is clearly past when they should've reordered. Say your product lasts about 30 days. Your reorder reminder handles day 25 to 35 or so. Winback would start somewhere after that, once they've missed the window. If you don't have a reorder reminder yet, build that first. It's the flow that stops people from lapsing in the first place.
The emails.
- The check-in. A short, plain reminder. "Running low?" with a one-click link back to the product they bought. No discount yet.
- The reason. Something new or something they missed. A new flavor, a bundle, reviews from people who use it the way they did.
- The offer. If you're going to use an incentive, this is where it goes. You've already caught the people who'd come back without one.
- The deadline. A reminder that the offer ends soon. Keep it short.
- The last call. One final note. Then hand anyone still unengaged to your sunset flow.
The spacing. I'd space these several days to a week apart. Winback isn't urgent like cart. Give each email room to land.
The exits. Anyone who places an order leaves the flow immediately. Nobody should get "we miss you" the day after they reordered.
Should the winback offer go in the first email?
I wouldn't. If you lead with the discount, you're paying everyone who would've come back with a simple reminder.
Save the incentive for email #3 or later. The early emails catch the people who just needed a nudge, at full price. The offer is there for the ones who need more.
If your margins are tight, try a winback without any discount first and see how far it gets you. You can always add one later.
How is winback different from a reorder reminder and a sunset flow?
These three get mixed up, so here's how I think about them.
- Reorder reminder: goes to customers right around when they should run out. It's there to stop the lapse before it happens.
- Winback: goes to customers who already missed that window. It's there to bring them back.
- Sunset: goes to subscribers who've stopped engaging at all. It gives them a last chance, then stops emailing them so your list stays healthy.
In a good setup they run in that order. Reorder reminder first, winback if that doesn't work, sunset at the end. I cover the last step in how to set up a Klaviyo sunset flow, and how all three fit into keeping customers buying in repeat purchase email strategy.
Common winback mistakes
These are the ones I'd check first:
- Stopping at one or two emails. The data says email #3 earns about 2x email #1.
- Leading with your biggest discount. You train customers to lapse on purpose.
- No exit on purchase. People who reorder keep getting "we miss you."
- Same flow for everyone. A customer who's ordered six times and one who ordered once aren't in the same spot. If you can, give your repeat buyers a slightly different message.
- No hand-off. When winback ends, the unengaged ones should move to sunset, not just keep getting campaigns forever.
Where these numbers come from
These are pooled numbers from 4 consumable brands, 2,897 campaigns and $17.7M in email revenue. It's a small sample, so treat it as directional, not a benchmark.
FAQ
How many emails should a winback flow have?
For a consumable brand, I'd start with four or five. In pooled data from 4 consumable brands, winback email #3 earned about 2x email #1, so cutting it at two leaves money behind.
When should a winback flow start?
After a customer has clearly missed the point where they should've reordered. That timing depends on how long your product lasts. Your reorder reminder should cover the window itself, and winback picks up after.
Should I offer a discount in my winback flow?
You can, but don't lead with it. Save it for the third email or later so the early emails catch people who'll come back at full price.
How long should a cart abandonment flow be?
Also longer than most people think. In the same data, cart email #5 still earned about $0.25 per person, the same as email #3.
Why should a welcome flow be short when winback should be long?
Because the money sits in different places. Welcome email #1 earns about 30x emails #2 to #5. In winback, the third email earns about 2x the first.
Founder of Bardown Digital, a Chicago-based Klaviyo email and retention agency for Shopify brands. He's worked with brands doing $500K to $100M a year, with $100M+ in email-attributed revenue across them.