Restaurant Lapsed Customers: What 85% With No Recorded Redemption Really Means

In a Feast production-data pull dated July 12, 2026, 29,805 of 34,952 opted-in guest records—85.3%—had no recorded redemption. Only 5,147 had a redemption on record.

That is a large gap in one Feast dataset. It is not an industry benchmark, and it does not prove that all 29,805 people never visited. Some may never have converted, some may have visited without a matched record, and the data does not give every guest the same amount of time after signup.

The number matters because it exposes a question most restaurants never ask: did the customer fail to visit, or did the restaurant fail to record the visit?

Do not call everyone a lapsed customer

A lapsed customer has completed a visit and then failed to return during a period when you expected another one. An opt-in with no verified visit is different.

Separate the list into four groups:

• people who opted in recently and still have time to visit;
• people whose visit window passed with no verified arrival;
• people who completed a first visit but have not yet had enough time to return; and
• people whose return window passed with no verified repeat visit.

Keep unresolved records separate. A response, reservation, or claim without a final visit outcome should not quietly become a no-show or failed conversion.

This distinction tells you whether the problem belongs to acquisition, the path from response to arrival, retention, or tracking. The same empty table can come from any of those causes.

First, find out why people never made the first visit

For opt-ins whose full visit window passed, ask a simple question before sending another discount.

Hi [first name], you signed up for [offer], but we do not have a completed visit recorded. What got in the way: timing, location, the offer, or something else?

In one Feast campaign audit, 37 replies were recorded after 51 lapsed-signup messages were sent. Guests mentioned timing, travel, and transportation. The audit did not establish how many messages were delivered or read, so the result should not be presented as a delivery-based reply rate.

The answers can still be useful. If timing dominates, clarify the eligible days and hours. If the offer is confusing, explain it in one sentence. If the guest expected that claiming the offer completed the process, improve the arrival instructions.

A reply diagnoses friction. It is not a visit.

Then repair one broken handoff

Follow the guest through:

Opt-in → response → reservation or claim → arrival → paid check

If people never respond, inspect delivery, audience, message, and timing. If they respond but do not book, inspect the offer or booking path. If they book but do not arrive, inspect confirmation, reminders, and cancellation. If they arrive but no check is connected, repair staff closeout and guest matching.

Choose one change based on the first weak stage. Sending a larger blast will not fix a booking problem, and a more generous offer will not fix missing POS records.

When the visit window closes, compare the number of eligible identified opt-ins with the number who completed a verified first visit. If the restaurant can connect those visits to paid, non-voided checks, report the linked gross sales separately. Linked sales are not final profit or automatic proof that the recovery message caused the visit.

Retention begins after the first visit

Only after a person completes a first visit should they enter a return cohort.

Give comparable first-time guests the same amount of time to return, then measure how many complete the next visit. Recent guests remain pending until their window closes. Unidentified visits and unresolved outcomes stay visible instead of becoming failures.

If first visits are healthy and returns are weak, the problem is retention. Use the first-timer-to-regular strategy to give the guest a specific reason for the next visit.

If too few people make the initial visit, a new loyalty reward will not repair the earlier gap. That is why opt-ins, first visits, and repeat visits must remain three separate results. For a concrete example, see what the FiiZ free-soda signups prove.

The 85% figure should change the questions you ask

The wrong conclusion is that 85% of restaurant customers always disappear. The right conclusion is that a large opt-in list can hide several different problems, and “redemption” alone does not tell you which one occurred.

For every campaign, keep these counts:

• eligible opt-ins;
• people whose visit window closed;
• verified first visits;
• first visits connected to paid checks;
• mature guests eligible to return; and
• verified repeat visits.

That is enough to see whether the next improvement belongs before the first visit, after the first visit, or inside the tracking itself.

Feast helps restaurants connect campaigns, guest follow-up, visits, and POS-linked sales so a large customer database becomes something operators can diagnose and act on.

Book a demo of Feast to find the gap between opt-in, first visit, and repeat visit.