The book generates more usable data than anything else in a restaurant except the POS, and almost all of it goes unread. Five numbers, half an hour a week, and the door stops being a matter of instinct — you find out whether the room is actually full or only feels full, whether your no-show habits worked, and how much you are paying for bookings that were always going to come.
📊 The five numbers
| Number | How to get it | The example restaurant |
| Seat utilisation | Covers served ÷ seats × turns the service allows | 120 of a possible 133 on a Friday — 90% |
| No-show rate | Covers that never arrived ÷ covers booked | 8%, about 50 covers a month |
| Average turn time | Seated at, gone at, by party size | 75 / 95 / 120 minutes |
| Channel mix and cost | Bookings by source, and what each one charges | 40% by platform, $1,200 a month |
| Turned away | The note from the door and the phone | About 30 groups a Friday |
The first one deserves an explanation because it is the one people get wrong. A four-hour service, with an average stay of an hour and a half, allows a fifty-seat room to sell about 133 covers. Selling 120 of them is 90% seat utilisation, which is genuinely good — and it also says the ceiling on a Friday is thirteen more covers, not fifty. Restaurants routinely plan as though a full room could hold far more than it can; this number is the reality check, and it is the one that tells you when the answer is no longer "book better" but "open another day" or "change the room".
🔧 What each number tells you to change
The point of a measurement is the decision behind it, so here is the map from number to action.
- Utilisation low with a full book? You are holding tables you never fill — the pacing caps are too tight, the slots do not match your real turn times, or bookings are being assigned to tables badly. Topic 3.
- Utilisation high and lots of people turned away? The room is genuinely the constraint. Now the conversation is about opening earlier, adding a seating, the terrace, or the table mix from the room course — and each of those has a cost that belongs to the budgeting course.
- No-show rate above about 5%? Confirmations first, phone numbers second, deposits for the big parties. Topic 4, in that order.
- Turn times drifting longer? Either service is slowing down — the kitchen course measures that — or the room got more comfortable, which may be exactly what you wanted. Both are fine; not noticing is not.
- Platform share rising? Your own channel is losing visibility. Move regulars back to your own link before the commission grows quietly.
- Lots of turn-aways at times you are not open? That is the cheapest expansion signal a restaurant ever gets, and it only exists if somebody wrote the noes down.
📅 Half an hour a week
Do it on the same day every week, with the person who runs the door, and keep it to the five numbers plus a look at next week's book.
Ten minutes reading what happened: utilisation on the two busy nights, no-shows marked, anything unusual. Ten minutes on next week: which slots are already tight, which are empty, whether a big party needs the room rearranged, whether anything is happening in town. And ten minutes changing exactly one thing — a cap, a slot length, the walk-in share, the confirmation message. One change at a time, so you can tell what it did.
Two habits make the meeting worth having. Write the numbers in the same place every week so the trend is visible without a report, and put the empty nights on the agenda too: a Tuesday at 40% utilisation is not a booking problem at all, it is a marketing one, and the courses that own it are waiting for exactly that signal.
🧭 What you have built
You decided whether the book is worth having, night by night, instead of inheriting one policy for the whole week (topic 1). You compared the three channels and found out what a platform really costs once you separate the diners it brings from the ones it merely intercepts (topic 2). You timed your own tables, turned them into slots with a buffer, and capped how many the kitchen can absorb in fifteen minutes (topic 3). You measured the no-show rate instead of guessing at it, and started converting silent absences into cancellations you can resell (topic 4). You built a waiting list that keeps people, learned to quote honestly, and turned "sorry, we're full" into a booking for another night (topic 5). And you know the five numbers that say whether the door is working.
With this the room is complete: the kitchen course fixed the flow behind the pass, the ambiance course fixed the room in front of it, and this one fills that room on purpose. What comes next in the library is the customer's own phone — ordering, loyalty and everything else that now happens in an app — and after that the part of the business that has nothing to do with service at all: what you buy, what you throw away and what the inspector asks for.