Restaurants don't have an acquisition problem. They have an amnesia problem, and they keep treating the symptom instead of the cause.
Read that imbalance again. The majority of the money walks in through people you've already met, and the majority of the spend goes on finding people you haven't. It's the commercial equivalent of ignoring your friends to impress strangers at a party.
Acquiring a new casual-dining customer costs roughly £100. If that guest hits one mediocre experience, an unrecognised allergy, a wobbly table nobody fixed, service that treats them like a stranger, they leave and never come back. That £100 is simply gone.
The instinct when revenue dips is to spend more on acquisition. Run another campaign. Offer another discount. But pouring new guests into a business that forgets them is pouring water into a leaking bucket: the faster you fill it, the more obvious the leak becomes.
The smarter play is unglamorous. Capture what your servers already know. Record the micro-preferences. Log the complaints nobody said out loud. Build a memory that outlasts a single shift, so the guest who came back a third time is met like someone you recognise, not processed like someone new.
Because loyalty was never a programme you bolt on. It's a pattern of behaviour, sentiment, and relationship strength, and you can't act on any of it if your data resets to zero every night. Fix the amnesia, and the acquisition problem you thought you had mostly disappears.
