Four people, one bottle, three hours, and a queue at the door. Why table turn is the number most venues never calculate, what it is actually worth, and the six things that change it without pushing anybody out.

Table nine has been occupied since half past nine. Four people, one bottle, three hours, and there is a queue at the door.

Nobody has done anything wrong. The table is simply worth less per hour than it could be, and almost no venue knows by how much.

The number nobody calculates

Revenue per table per hour = table spend ÷ hours occupied

⚠️ Most venues measure spend per table and never divide by time, which makes a three-hour table look identical to a one-hour table of the same value.

Why that matters on a full night

On a quiet night, occupancy is free — the table would otherwise be empty. On a full night, every hour a table is held is an hour it cannot be sold to somebody at the door.

Quiet night   — turn is irrelevant
Full night    — turn is the entire constraint

⭕ The same policy cannot be right on both nights, and most venues run one policy.

The three-hour table is not the problem

⚠️ A three-hour table spending well is excellent. The problem is the three-hour table spending like a one-hour table.

What actually stops a table turning

Six things, and only one of them is the guests.

One: nobody has offered anything for forty minutes

The most common cause by a distance. The table would happily order again and nobody has asked.

⇒ The First Fifteen Minutes Of A Shift

Two: the bill takes twenty minutes to arrive

A group ready to leave at midnight leaves at twenty past, holding the table for twenty minutes at zero value.

⇒ The Split Check At A Big Table

Three: the group is waiting for one person to come back

Frequently from outside.

⇒ The Smoking Area Nobody Manages

Four: the table was never cleared

A table with empty glasses on it reads as still occupied to the floor and as finished to the guests. Both wrong.

Five: nobody knows the table is free

It cleared eight minutes ago, the door has a queue, and the two facts have not met.

⚠️ This is a communication gap, not a capacity problem, and it is invisible in every report.

⇒ The Board In The Dressing Room

Six: the guests genuinely want to stay

The only one of the six that is about the guests, and the only one where pushing is a mistake.

What to do about each

For one, two, four and five: fix the process

None of these require any conversation with a guest.

A check-in on every table at a set interval
The bill offered, not waited for
Clearing as a continuous job, not an end-of-table job
One person who knows which tables are free

⭕ Four changes, no policy, no awkwardness.

For three: make coming back easy

⇒ Getting Guests Home At Three

For six: decide in advance

⚠️ A minimum, a time limit, or nothing — all three are legitimate. Improvising at eleven o'clock is not.

⇒ Enforcing A Minimum Without Losing The Table

Measuring it without a system

You need one busy night and a piece of paper.

What to record

Table number, time seated, time cleared, spend

⭕ Four fields, one line per table, filled in by the host as it happens.

What it gives you

Average occupancy in hours
Revenue per table hour
The spread — which is the interesting part

⚠️ The average hides everything. Two tables at three hours and six at one hour average out to something that describes neither.

⇒ The Number You Check First

The number that starts the conversation

Your best table hour ÷ your worst table hour

⇒ If the ratio is large, the opportunity is in the bottom half, not in adding seats.

What venues get wrong

Adding capacity instead of using it

More tables — costs money, needs space, needs staff
Better turn — costs nothing

⚠️ Venues frequently consider a refit before they have measured whether the room they have is being used.

⇒ What Your Second Location Actually Costs

Pushing people out

⛔ It is visible, it is remembered, and it costs more in returning guests than it gains in one extra sitting.

⭕ Every fix on the list above is invisible to the guest. That is the point.

Treating every night the same

A policy that helps on Saturday hurts on Tuesday, where the goal is the opposite: keep them as long as possible.

⇒ The Slow Tuesday You Keep Open

The door side of the same number

Turn only matters if somebody is waiting. Which means the two numbers have to be looked at together.

Tables held × hours held — on a night with a queue
People turned away that night

⚠️ A venue that turns people away while running three-hour tables at low spend is losing twice, and neither loss appears anywhere.

⇒ The Night You Turned People Away

And the waiting list

A queue with no list is a queue that walks. People who would have taken a table at 23:20 leave at 23:05 because nobody told them anything.

⇒ The Waiting List You Do Not Keep

What good looks like

The forty-minute test

Walk the floor at peak and find the table that has been offered nothing for longest.

Under 20 minutes — working
Over 40          — that is your revenue, sitting there

The clearing test

Count tables with empty glasses on them at peak. Any number above two is a clearing problem.

The free-table test

Ask the door how many tables are free right now. Then walk the floor and count.

⚠️ The two numbers are rarely the same, and the gap is money.

The booking that holds a table it does not use

A reservation is a table taken out of circulation before anybody sits at it, and most venues do not count that time.

The hour before they arrive

Table held from 21:00, party arrives 21:40
40 minutes of a Saturday table, sold to nobody

⚠️ A fifteen-minute grace is a policy. Forty minutes is an accident — usually because nobody was watching the clock on that table.

⭕ Set a hold time, and give the floor permission to release it without asking a manager.

The party of eight that arrives as five

The table was sized for the booking, not for the people. Three seats sit empty all night, and the group at the door who needed five went elsewhere.

⇒ The Waiting List You Do Not Keep

The no-show nobody chased

⛔ A booking that does not arrive and is never confirmed costs the full evening value of that table.

⭕ A confirmation the day before costs one message and recovers most of it.

The bottle service table is a different question

Where a table carries a spend minimum, turn stops being the measure and the arithmetic changes completely.

The minimum is the floor, not the target

Minimum met at 22:30, table held until 02:00

⇒ The table has done its job and is now producing nothing. That is a service opportunity, not a policy problem.

⇒ Enforcing A Minimum Without Losing The Table

Two tables at the same minimum are not equal

One reaches it in an hour and orders again. One reaches it at closing. Same revenue, very different nights — and a report showing only spend cannot tell them apart.

What to measure instead

Time to minimum ／ spend after minimum

⚠️ The second number is where the entire opportunity sits, and almost nobody has it.

What to change first

Three tests above, six causes, and limited attention. Order matters.

Start with clearing

It is the cheapest, it needs no conversation with anybody, and it is usually the largest single cause.

Then the bill

⭕ Offering the bill rather than waiting to be asked costs nothing and returns twenty minutes per table on a busy night.

Leave policy until last

⚠️ Minimums and time limits are the most visible change and the smallest return. Every venue reaches for them first.

Three to hold

1  Spend per table means nothing without hours occupied
2  Five of the six causes are process, and invisible to the guest
3  Turn matters only when somebody is waiting — look at both numbers together

Where the record has to sit

Seated, cleared, spend — four fields per table, kept with the shift record. One busy night of it tells you more about your capacity than a year of gross takings.

⇒ The Report You Open Every Monday

The Table That Never Turns