Loyalty & Retention

How to Win Back Lapsed Café Customers (Without Discounting to Death)

By The Tany Team 9 min read

The most overlooked growth lever in an independent café is not finding new customers — it is noticing when an existing one quietly stops coming in, and doing something about it before they are gone for good. Acquiring a brand-new customer is estimated to cost roughly five to seven times more than retaining one you already have, and a small lift in retention compounds hard: research widely cited in the industry suggests a 5% increase in retention can raise profits by 25% to 95%. Yet most cafés have no system at all for the regular who used to come every Tuesday and hasn’t been seen in a month.

This guide is a practical, margin-aware playbook for winning lapsed customers back. It assumes you are an independent café or restaurant, probably on Square, and that you would rather send the right message to the right person than blast a 50%-off coupon at your entire list.

What does “lapsed” actually mean for a café?

“Lapsed” is not a fixed number — it depends on how often your customers normally visit. A useful definition: a customer is lapsed when they have gone 1.5 to 2 times longer than their personal normal gap without coming in.

  • A daily regular who skips a week is a yellow flag.
  • A weekly regular silent for two to three weeks is worth a nudge.
  • A monthly customer is not lapsed at five weeks — that is just their rhythm.

If you cannot compute per-customer intervals yet, start with a single blunt window: no visit in 30 days for a café, or 45–60 days for a restaurant people visit less often. It is imperfect, but a blunt rule that ships beats a perfect rule that never does. Tune it after you see who responds.

The reason precise timing matters: the longer someone is gone, the harder and more expensive they are to recover. A nudge at week three is a friendly reminder. A nudge at month six is a resurrection, and usually needs a real incentive to work.

Why win-back beats chasing strangers

The economics are lopsided, and worth stating plainly because they justify spending your limited time here instead of on ads.

FactorNew-customer acquisitionWinning back a lapsed regular
Relative cost5–7x more expensive (some studies cite up to 25x)A fraction; you already have their contact info
They know you?No — you pay to build awareness and trustYes — they already liked you enough to come back before
Spend per visitFirst-timers spend less on averageRepeat customers are reported to spend meaningfully more per order than first-timers
Channel costPaid ads, marketplace commissionsA push, an email, or a single text

Industry data also suggests a large share of restaurant revenue comes from repeat guests, and that a majority of first-time diners never return. That second fact is the warning: customers leak constantly and silently. A win-back system is simply a net under the leak.

For a fuller picture of what each repeat visit is worth over time, see our breakdown of customer lifetime value for a coffee shop — the win-back math only makes sense once you know what one recovered regular is worth.

Step 1: Make sure you can actually see lapsed customers

You cannot win back a customer you never identified. The prerequisite for everything below is customer data tied to visits — which means moving people off anonymous cash and onto something that records who they are.

On Square, customer profiles get created and enriched when guests:

  • Join your loyalty program (the highest-signal source for a café).
  • Order through online ordering or a branded app with an account.
  • Save a card on file or get an emailed/texted receipt.

Square’s Customer Directory then builds automatic segments, including a Lapsed group — customers who haven’t returned within a period you define. If your directory is mostly empty, that is the first problem to fix; a loyalty program is usually the fastest way to populate it, and we cover the options in what a coffee shop loyalty program really costs.

Step 2: Segment — not everyone is worth the same effort

Resist the urge to treat all lapsed customers identically. Split them into three buckets and spend accordingly:

  1. High-value lapsed — used to come often and spend well. These are worth a personal, possibly incentivized message. Losing one is genuinely expensive.
  2. Casual lapsed — came occasionally. Worth a cheap, automated nudge, not a discount.
  3. One-and-done — visited once, never returned. Low odds; include them in a broad automated campaign but spend nothing extra.

This segmentation is what keeps win-back profitable. A blanket “20% off, come back!” to your whole list hands a discount to people who would have returned anyway and to people who will never return — both pure cost. Targeting the high-value bucket with care and the rest with near-free automation is the entire game.

Step 3: Pick the channel you own

How you reach lapsed customers determines both cost and reach. The three realistic channels:

  • Push notifications (via your own app): effectively free per send once the app exists, instant, and impossible to miss on a lock screen. Limited to people who installed your app.
  • Email: very cheap, good for longer messages and images, but lower open rates and easy to ignore.
  • SMS: highest reach and read rate, but you pay per message and must respect consent rules.

There is no single right answer — it is a cost-versus-reach trade, and we lay out the full comparison in email vs SMS vs push for café marketing. The strategic point: a channel you own (your app, your list) costs nothing per recovered order, whereas re-acquiring the same person through a delivery marketplace or paid ad costs every time. Owning the relationship is the whole reason win-back is cheap.

Square Marketing can automate this directly: it sends scheduled email and text campaigns to segments like Lapsed and Birthday, with email priced by audience size (roughly $15/month for up to 500 customers, scaling up from there) and text priced by message volume. That is a fine starting point. A branded app with push removes the per-message cost entirely, which matters once your list grows.

Step 4: Write a message that protects margin

The message itself decides whether you win the customer back at zero cost or buy them back at a loss. A simple ladder, cheapest first:

Tier 1 — the plain reminder (no discount).

“We’ve missed you at [café]. Your usual oat latte is two taps away — order ahead and skip the line.”

A surprising share of lapses are just broken habit, not dissatisfaction. A specific, friendly nudge that names their usual order recovers these for free. Always try this first.

Tier 2 — a value-add, not a markdown. Offer something that costs you little but feels generous: double loyalty points this week, a free size upgrade, or a pastry with any drink. These protect your headline price and your perceived value, which a percentage-off coupon erodes.

Tier 3 — a real, capped offer (high-value lapsed only). If a high-value regular ignores tiers 1 and 2, then spend money: a fixed-dollar credit (“$5 toward your next visit”) is safer than a percentage because it caps your downside and nudges a real order. Make it expire — urgency is what converts a coupon into a visit.

A worked example to keep this honest (the numbers below are illustrative, not a measured Tany result): say you have 200 lapsed customers and a $6 average ticket. A free push reminder that brings back even 10% is 20 recovered visits at roughly $120 in sales — at essentially no marketing cost. The same 10% won with a $5 credit costs you $100 in credits against that $120, leaving almost nothing. That gap is exactly why you lead with the free nudge and reserve discounts for customers whose lifetime value justifies them.

Step 5: Make the return effortless

Winning attention is wasted if coming back is a hassle. The message should land on a one-tap path to order: order-ahead pickup so they skip the line, their order history so re-ordering “the usual” is instant, and saved payment so checkout is frictionless. Every extra step between the notification and the coffee leaks the customers you just re-earned.

This is where a branded order-ahead channel pays off: the same app that sends the free push also stores their usual, their loyalty balance, and their card — so the path from “we miss you” to “order placed” is seconds, not a re-introduction.

Step 6: Measure and tune the window

Treat win-back as a loop, not a one-off. Each month, check: how many customers entered the lapsed segment, how many a campaign brought back, and what it cost. Then adjust your lapse window and your offer tiers. If almost everyone you flag at 30 days comes back on a free nudge, your window may be too tight (you are nudging people who weren’t really gone). If few respond, you are likely catching them too late — tighten the window.

Where a branded app fits

You can run a competent win-back program on Square Marketing alone, and many cafés should start exactly there. The ceiling you eventually hit is cost and ownership: every SMS costs money, email opens are middling, and the relationship still lives partly in someone else’s tool.

A branded app on your existing Square POS raises that ceiling — unlimited free push to re-engage lapsed guests, their saved usual and loyalty balance in one place, and a one-tap path back to ordering. That is the niche Tany fills: a branded iOS and Android order-ahead app with self-running loyalty and push, live in about a day on your Square POS for $99 CAD/month per location. It is one way to own the win-back channel, not the only one — the playbook above works regardless of the tools you choose.

The takeaway: stop treating lapsed customers as gone. Define the window, watch the segment, and send one timely, specific, margin-aware message before the habit fully breaks. It is the cheapest growth you have.

Sources

Frequently asked questions

When should I consider a café customer lapsed?
A customer is lapsed when they have gone meaningfully longer than their normal gap between visits without coming in. A practical rule is 1.5 to 2 times their typical interval: if a regular usually visits weekly, flag them at roughly two to three weeks of silence. For a café with mostly daily and weekly regulars, a 30-day no-visit window is a reasonable starting trigger you can tune over time.
Do I have to offer a discount to win customers back?
No. Discounts are the easiest lever but the most expensive one. Many lapsed customers simply forgot or fell out of routine, and a timely, personal reminder — 'we miss you, your usual is one tap away' — recovers a share of them at zero margin cost. Reserve real discounts for higher-value customers who do not respond to a plain nudge, and cap the offer so it never costs more than the visit is worth.
What is the cheapest way to message lapsed customers?
A push notification through your own app is the cheapest per-message channel because there is no per-send fee once the app exists, unlike SMS which carries a cost per text. Email is also low-cost. The trade-off is reach: push only reaches people who downloaded your app, SMS reaches the most people but costs the most, and email sits in between on both cost and open rate.
How do I find lapsed customers if I only use Square?
Square's Customer Directory groups customers into automatic segments including Lapsed (customers who have not returned in a set period), and Square Marketing can send an automated email or text campaign to that segment on a schedule. You need customer data flowing in first, which happens when guests check out with loyalty, online ordering, or a saved card rather than anonymous cash.