Most restaurant marketing budgets are spent finding people who have never visited. Meanwhile the guests who already came, enjoyed it, and would happily return are contacted by nobody.
That is backwards, and it is backwards in a measurable way.
A new guest costs whatever your acquisition channel charges — aggregator commission, paid social, a discount to get them through the door. A returning guest costs the price of a message.
Returning guests also tend to spend more per visit, because they are not spending the first twenty minutes working out what kind of place it is. They order with more confidence, try more, and are far more likely to order a second course.
Not unwillingness — data. Most venues have no idea who was in last week. The card machine knows a transaction happened; it does not know who, what they ordered, or whether they had been in before.
Without that, retention marketing is impossible. You cannot bring back a guest you cannot identify.
Once visits attach to a guest, useful questions become answerable. Who used to come fortnightly and has not been in for two months? Who only ever comes at lunch and might be worth a dinner offer? Who spends well and should be recognised as a regular before they have to say so?
Each of those is a different message. Blanket discounting cannot do any of them, and blanket discounting is what most venues fall back on.
The instinct is to bring guests back with money off. It works, and it trains your best guests to wait for a discount before booking.
Recognition is cheaper and more durable — knowing a guest's usual table, remembering an allergy without being told, moving them into a tier because of what they actually spend. Guests notice being known far more than they notice ten percent.
The fastest retention win is not a loyalty scheme. It is a list of guests who used to come regularly and stopped. They already liked the place. Most of them have no particular reason for stopping, and a well-timed message brings a meaningful share of them back.