Win back customers before they're gone
A customer who bought once and went quiet is cheaper to re-activate than a stranger is to acquire, they already trust you. The move is to define 'lapsed' by your real reorder cycle, run a staged winback, and sunset the truly dead before they drag deliverability.
AI can build the flow and the segments from your data. You set the timing to your cycle, because only you know it.
Fastest path: one prompt, end to end
🤖 AI prompt — paste into ChatGPT / Claude
You are a retention strategist. Use MY numbers only.
Product + typical reorder cycle: [e.g. coffee ~30 days]
Customer counts if I have them: bought once then quiet [#], lapsed 2x+ buyers [#]
Contribution margin %: [%] Best incentive I can afford: [% or $]
Do this:
1. Define 'lapsed' for MY reorder cycle (e.g. no order in ~2x the cycle), not a generic 90 days, and explain why cycle-relative beats fixed.
2. Design a staged winback flow: (1) a value/reminder nudge with no discount, (2) social proof / what's new, (3) a time-boxed incentive as the last step, not the first. Give timing + message purpose for each.
3. Add a sunset rule: when to stop emailing a truly-dead contact to protect deliverability (ties to the email deliverability note).
4. Recommend measuring reactivation rate, and warn against leading with the discount (it trains people to lapse).
If a number is missing, ask; do not guess.
Output: lapsed definition + staged winback flow + sunset rule + metric.
Or do it in 4 steps
- Define 'lapsed' by your cycle, not a calendar. A coffee buyer silent for 60 days (2x a 30-day cycle) is lapsing; a furniture buyer at 60 days is normal. Cycle-relative timing is what makes the winback land at the right moment instead of nagging or arriving too late.
- Lead with value, not a discount. Start the winback with a reminder, a what's-new, or genuine social proof. Opening with a coupon trains customers to lapse on purpose to earn one, save the incentive for the last step.
- Make the incentive time-boxed and last. If value nudges don't re-activate them, then a limited-time offer as the final touch. The deadline creates urgency; the fact that it's earned only after softer touches protects your margin and your full-price norm.
- Sunset the truly dead. Someone who ignores the whole winback should be suppressed from regular sends, mailing dead addresses drags your deliverability down for everyone (see the deliverability note). Winning back and knowing when to let go are the same discipline.
Worked example (labeled): coffee, 30-day cycle. Winback fires at day 45 (no discount, 'running low?'), day 55 (a new roast + reviews), day 65 (10% off, 72 hours).
A buyer who'd have churned reorders on the reminder alone, at zero incentive cost; only the genuinely hesitant ever see the discount. Contrast a 'here's 20% off, come back!' blast on day 30, which discounts people who'd have reordered anyway. Time it to your cycle and lead with value.
Track reactivation rate and keep the sunset rule running; a winback that leads with discounts trains churn instead of curing it.