Retention or acquisition — where should an F&B operator spend next?

This is usually argued as a philosophy question and it is actually an arithmetic one. There is a number that decides it: your rolling 90-day repeat rate. Below roughly 35%, acquisition spend leaks out as fast as it comes in and retention is the only sensible investment. Above it, the leak is under control and acquisition starts compounding.

The cost asymmetry is large and well established. Bringing in a new F&B customer — advertising, marketplace commission, launch discount, the free item that got them through the door — costs four to seven times what it costs to bring back someone who already knows the venue. The lapsed customer needs no education, no trust-building and no introductory discount; they need a reason and a reminder.

The trap is that acquisition is visible and retention is not. A paid campaign produces a dashboard with impressions, clicks and new customers. Retention produces an absence of loss, which no dashboard shows by default. So operators keep buying the visible thing, and the leak is only diagnosed when acquisition costs rise or a competitor opens nearby — at which point the venue discovers it has been running on replacement for two years.

But retention is not unlimited. Once your repeat rate is strong and your regulars are visiting at their natural frequency, additional retention spend has nowhere to go — you cannot make a daily customer come twice daily. Past that point growth genuinely does require new customers, and the right move is to shift the ratio deliberately rather than to keep optimising a maxed-out lever.

A practical allocation rule: under 25% repeat rate, put nearly everything into retention and enrolment. Between 25 and 40%, split roughly 60/40 toward retention. Above 40%, invert it — the retention machine is working and acquisition now compounds because new customers actually stick. Recheck quarterly; the number moves.

One caveat worth naming: retention work has a prerequisite that acquisition does not. You cannot retain customers you cannot identify. If you are capturing a single-digit percentage of transactions, the first spend is not on retention campaigns — it is on enrolment, which is the cheapest line item in this entire discussion and the one that unlocks the rest.