A regular — call her Maria — comes in most Thursdays. She booked the table under her partner's name. She paid on a card the till has never tied to her. She joined the WiFi with a Gmail address she uses for nothing else. She gets your marketing emails at a third address she gave you eighteen months ago. To you, she's Maria: your Thursday regular who likes the corner table and never finishes a starter. To your software, she is four different strangers.
That gap — between what you know and what your systems know — is the most expensive thing in your venue. Not the rent. Not the stock. The fact that the booking system, the till, the WiFi and the marketing tool each hold a partial, slightly wrong, never-reconciled picture of the same person, and none of them will ever introduce her to the others.
Fragmentation has a cost, and you pay it every service
The host can't see she's a regular when she walks in, because the booking is under another name. So she gets treated like a first-timer — politely, but like a stranger.
Your marketing can't tell that the person who hasn't been in for six weeks is the same person who books every Thursday, because the lapsing-guest list and the bookings live in different tools. So the win-back message never sends — or worse, it sends to someone who was in last night.
Your end-of-month numbers can't tell you that a Thursday regular on £40 a head is worth more over a year than a Saturday walk-in on £80, because spend lives in the till and frequency lives in the booking system and the two have never met.
You don't notice this cost because it never arrives as a bill. It arrives as the regular who quietly stopped coming, and you never knew why.
One record changes the arithmetic
The fix isn't another tool. It's the opposite — it's making every tool point at the same person. One guest record: the booking, the cover, the card, the WiFi login, the marketing consent, the loyalty balance, the allergy note, all attached to Maria, because they are all Maria.
When that's true, ordinary things become possible that were impossible before. The host screen shows you she's a regular before she's through the door, with her last visit and the fact she's coming up to a birthday. The "we've missed you" message goes only to the people who have actually lapsed, and stops the moment they book. And you can finally answer the question every operator should be able to answer and almost none can: which of my guests are worth keeping, and are they coming back?
None of that needs an AI, or a data team, or a consultant. It needs one record instead of four.
Why this is hard to buy, and easy to fudge
Plenty of software will tell you it does this. The honest test is to ask where the guest record actually lives. If your loyalty is a separate app, your bookings a separate platform, your till a separate box and your marketing a separate subscription, then "one guest record" is really a sync job between four vendors who would each rather be the one you keep. Syncs drift. Duplicates breed. The record is only ever as good as the worst integration in the chain.
The only way one guest record stays true is if the booking, the till, the WiFi and the marketing were built to write to it from the start — not bolted together afterwards.
The question to ask first
That's the bet Grace is built on, and I won't pretend to be neutral here: I built Grace because I was the operator with four systems and four versions of Maria, and I got tired of treating my own regulars like strangers.
But you don't need Grace to take the point. Whatever you run, ask it the simple question: when Maria walks in on Thursday, does anything you own actually know it's her? If the answer is no, that's the thing to fix first. The marketing, the loyalty, the reporting — all of it only works once it does.