Winning back customers who quietly stopped buying

Churn in ecommerce is silent. Nobody cancels. They simply do not come back, and because there is no event, there is nothing to react to — the loss is invisible in every report that counts what happened rather than what stopped happening.

Meanwhile the lapsed customer is the cheapest sale available. They know you, they have handed over card details before, and they liked the thing enough to keep it.

Enrolment happens on its own

Customers who cross the line into inactive are enrolled automatically, on a schedule, using the same behavioural segments as everything else. That is the whole reason it works: a manual win-back is something you do in January and then never again, and the customers who lapse in March are lost to inattention rather than to competition.

Pick the window from your own repeat cycle

“Inactive” is meaningless in the abstract. Ninety days is alarming for coffee and unremarkable for garden furniture.

Set it from your actual repeat interval — roughly twice the typical gap between orders is a sound starting point. Too short and you are pestering people who were going to buy anyway, which is worse than doing nothing because it trains them to ignore you. Too long and you are writing to someone who has already found a replacement.

Discount last, not first

The reflex is to lead with money off. It works, and it also teaches your customers that waiting produces a discount — which converts a full-price buyer into a discount-only buyer permanently. You have not won them back; you have repriced them.

Try the cheaper things first. What is new since they last looked. What they bought before, back in stock. Something genuinely useful about the product they own. Keep the discount for the people who ignored all of it, and accept that some of them are simply gone.

Knowing when to stop

A win-back sequence needs an end. Somebody who has ignored four messages is telling you something, and the fifth costs you deliverability across your whole list, not just with them.

Being willing to stop writing to people is what keeps your email arriving in inboxes for everyone else. This is the part most automation gets wrong, because sending more is always the easier option.

Measure against a holdout

The uncomfortable question about any win-back: how many of those customers would have returned anyway? Attribute all of it to the campaign and you will believe it works far better than it does.

Hold back a slice of the segment and send them nothing. The difference between the two groups is what the campaign actually produced. It is the only honest measurement, and if the difference is small the answer is a better offer rather than more sending.