You know that customer who used to come in every morning? The one who always ordered a flat white and knew your barista by name? Then one day they just... stopped. It happens to every coffee shop, and it's more common than you think.
Why customers lapse
People don't usually leave because they had a bad experience. Life just gets in the way. They change their commute. A new place opens closer to home. They start working remotely on Wednesdays. The reasons are mundane, but the effect is real.
The important thing to understand is that most lapsed customers aren't actively avoiding you. They've just fallen out of the habit. And habits, unlike grudges, can be rebuilt.
Timing is everything
The worst time to reach out to a lapsed customer is six months after they've stopped coming. By then, they've probably found a new routine. The best time is much sooner, within two to three weeks of their last visit.
This is where data becomes your friend. If you can see that a customer who normally visits every five days hasn't been in for fifteen, that's a signal. Not an alarm. A signal. It means a gentle nudge might be all they need.
"Lapsed" is a different number for every customer
This is the part most schemes get wrong, and it is the difference between a useful list and a list nobody acts on.
A single fixed rule, thirty days for everyone, flags the wrong people in both directions. Thirty days is nothing for a customer who has always come once a month, so you will pester somebody who is behaving perfectly normally. And thirty days is an eternity for the Tuesday-morning regular who came every week for a year, so by the time they appear on your list they have had four weeks to form a new habit somewhere else.
The definition that works is relative to the person: how long since their last visit, compared with their own usual gap. A five-day customer at fifteen days is at three times their normal interval and that is a genuine signal. A thirty-day customer at forty-five days is at one and a half times theirs, and is probably just having a quieter month.
You can do this by hand for the twenty regulars you could name from memory. Beyond that it needs something that remembers each customer's own rhythm, which is really the argument for measuring any of this at all.
What to say (and what not to)
The message matters as much as the timing. Nobody wants to receive an email that says "We miss you! Come back!" It feels desperate and generic.
Instead, lead with value. If the customer is close to completing their stamp card, tell them. "You're one stamp away from a free coffee" is specific, useful, and gives them a reason to act. It's not about guilt. It's about reminding them of something they've already invested in.
If they've already earned a reward but haven't redeemed it, that's even better. "You've got a free coffee waiting" is hard to ignore.
The segments that matter
Not all lapsed customers are the same. Someone who visited twice and disappeared is very different from a former regular who's been away for a month. Your approach should reflect that.
For the purposes of a coffee shop, three segments are usually enough:
At risk. Regulars sitting at roughly twice their own normal gap. These are your highest-value targets by a distance. They know you, they liked you enough to establish a rhythm, and they have not yet replaced it with another one. Almost all of the value in this exercise is here.
Lapsed. Three or four times their normal gap. The habit has gone and you are asking them to start again rather than continue. A waiting reward is the strongest thing you have, because it is a reason to return that belongs to them already.
Gone. Many times over, or months with no pattern to compare against. Be honest with yourself: most of these aren't coming back, and the effort is better spent stopping the at-risk group from joining them. The one exception worth making is a customer sitting on an unclaimed reward, because you owe them that regardless of whether they ever spend again.
Making it automatic
The beauty of a system that tracks visit patterns is that you don't have to manually monitor every customer. Set up your segments, craft two or three well-written email templates, and let the system identify who needs a nudge and when.
The goal isn't to bombard people with messages. One well-timed email every few weeks is far more effective than a weekly newsletter nobody reads. Quality over quantity, the same principle that probably applies to the coffee you serve.
How to tell whether any of it worked
One number, and it is not opens or clicks.
Of the people you contacted, how many walked back in within a fortnight? That is the whole measurement. An email that gets opened by half its recipients and brings nobody through the door has done nothing except use up your welcome, and the open rate will cheerfully hide that from you for months.
Two things worth doing alongside it. Leave some people out on purpose the first time or two: if a similar share of the people you did not contact came back anyway, the nudge is taking credit for customers who were returning regardless. And watch whether the ones who did return stayed returning, or came once and vanished again, because a message that buys a single visit is a different and much weaker thing than one that restarts a habit.
The bottom line
Customer retention isn't about preventing people from ever leaving. That's impossible. It's about shortening the gap between when someone drifts and when you notice. The shops that do this well don't just keep more customers. They build the kind of steady, predictable revenue that makes everything else easier.