A customer win-back email workflow serves former buyers who are beyond their usual purchase cycle and still eligible for contact. Review repeat-order intervals by product group, then offer a relevant reason to return. Do not start by sending every previous customer the same discount after the same delay.
Define inactivity around the natural buying cycle
Inspect repeat customers’ order dates by product group. Do not count split same-day orders or cancellations as genuine repeat purchases. Pack size, seasonality and first-time versus returning customers can affect the interval. If data is sparse, describe your initial timing as an assumption rather than established behaviour.
Klaviyo’s win-back guide also ties timing to the buying cycle. A regularly consumed cleaning product and a table used for years need different definitions of inactivity. A replenishment reminder may suit the former, while care information or a relevant new product may suit the latter. This is a planning example.
Do not put every customer into the same message
A customer with an unresolved support request differs from someone happily using a long-lasting product. Check whether the last order completed, whether a return is open and whether communication preferences permit the message. Reflect those answers in entry rules.
| Customer situation | Relevant approach | Check |
|---|---|---|
| A regularly used product may be needed again | A concise, useful reminder | Does the actual replenishment cycle fit? |
| A durable product was purchased | Care guidance or a relevant development | Is repeat-purchase pressure unnecessary? |
| An issue or return remains open | Service and resolution first | Is marketing overtaking the problem? |
| Preferences have changed | Respect the current preference | Is sending still appropriate? |
| A new purchase occurred | Exit the win-back workflow | Has the latest order data arrived? |
Provide a meaningful reason to return
Sample subject: “Current options for the product you use”. Explain which options genuinely relate to the earlier purchase and why. For a durable product, a verifiable benefit such as an updated care guide may be more useful than trying to sell the same item again. Do not promise an update that has not happened.
If offering a discount, state its eligible products and real expiry. Do not re-enrol unsubscribed people under the label of win-back. Check stock, compatibility and destination links; “we miss you” is not a substitute for a reason to contact someone.
Plan timing and exit together
Enter a test customer, create a new order before the first message, and verify that later messages stop. Repeat the test for an unsubscribe and an open support case. Record when the source data arrives: a workflow using stale status can still send the wrong message.
Document the maximum number of messages and the re-entry condition. Do not leave non-responders in an endless cycle. Review total frequency alongside other active campaigns.
Review commercial and experience outcomes together
Consider repeat purchasing, useful replies, renewed resource use and unsubscribes. An attributed order does not itself prove the message generated an additional order. Where that distinction matters, investigate it through an appropriate comparison design.
Test records that should be excluded too: recent purchasers, unresolved issues, changed preferences and customers whose normal buying interval has not passed. A useful win-back plan understands who should not receive a message now as well as who should.