The First Twenty Minutes Decide The Bill

Two tables, same size, same night, same hour. One spends 20,000 and leaves at midnight. The other spends 60,000 and stays until two.

The usual explanation is that they were different customers. Sometimes. More often, they were the same kind of customer and the first twenty minutes went differently.

The short version

• The spending pattern for a table is usually set before the second drink
• Three things decide it: how fast someone arrives, who arrives, and whether a second round is ever offered
• A table left alone for ten minutes rarely recovers to a high bill
• The measurable version is time from seating to first entertainer
• Long waits cluster in the same two places every night, and they are predictable

Why the window is so short

A group arriving decides, quite quickly and without discussing it, what kind of night this is going to be.

In the first 20 minutes	What the table concludes
Seated, drinks taken, nobody comes for 12 minutes	"We'll have one and see"
Seated, someone over within 3 minutes	"We're settled in"
First drink arrives, nobody offers a second	"We'll go somewhere else after this"

⚠️ Nobody at the table says any of this out loud. It shows up an hour later as a short bill, and by then it is read as a quiet night.

One — measure the wait, because you cannot feel it

Ten minutes on a busy floor feels like two to staff and like twenty to a waiting table.

Time from seating to first entertainer at the table

Under 3 minutes …the table settles
3 to 7          …acceptable
Over 10         …the ceiling on that bill is already set

⇒ Record it for two weeks. You do not need every table; ten a night is enough to find the pattern.

Where the long waits cluster

One  the first 40 minutes after opening, before the floor fills
Two  the changeover when one rotation ends and the next has not started

⚠️ Both are structural, not effort. Nobody is slacking; the rota simply has a hole in it, and the same hole appears every night.

Two — the changeover hole

This is the most common and the easiest to fix.

Rotation ends at 10:20. Next starts at 10:20.
In practice: girls leave the floor at 10:18, the next are ready at 10:26
⇒ an eight minute hole, twice a night

⇒ Any table seated in that window gets the long wait.

The fix

Stagger the rotation so half change at 10:20 and half at 10:35

⇒ The floor is never empty, and the changeover stops being visible from a table.

⚠️ This costs nothing. It is a scheduling change, not a staffing one.

Three — who arrives matters as much as how fast

A fast arrival by the wrong person is not much better than a slow one.

A group of four, and one entertainer arrives
⇒ she cannot hold four people. The table stays polite and spends once

Send to match the group size where you can

⇒ Two to a four-top beats one, even if it means the next table waits an extra minute.

⚠️ Splitting thin across every table is the same mistake as spreading a staff shortage evenly. Some tables are worth covering properly and some can wait ninety seconds.

Which tables to cover first

One  larger groups
Two  tables that asked for someone by name
Three anyone who has been in before
Four  everyone else

Four — the second round is a moment, not a policy

Most tables that stop at one drink were never asked.

Weak  …wait until the glass is empty, then ask
Good  …ask when the glass is about a third left

⇒ At empty, the group has already had the "shall we go?" thought. A third left is before it.

"Shall I bring the same again?"

⚠️ Naming the same drink is easier to say yes to than an open question. "Anything else?" invites a decision; "the same again" invites a nod.

The number that tells you it is not happening

Second drinks ÷ first drinks, by night

⇒ If it is low and the wait times are fine, the offer is the gap, not the service.

Five — what not to do with this

Two ways this goes wrong when owners act on it.

Rushing the floor  …tables notice being processed, and it reads worse than waiting
Scripting it       …a memorised line lands worse than no line

⚠️ The fix is structural — the rota hole, the group matching, the timing of the offer. It is not "try harder", and presenting it that way to the floor will make it worse.

Six — the numbers to keep

One   time from seating to first entertainer, sampled
Two   bill by table against that wait time
Third second drinks as a share of first
Four  bills for tables seated during changeover, versus other tables

⚠️ Four is the one that proves it to a sceptical floor manager. If changeover tables have visibly lower bills, nobody argues about the rota after that.

A sheet for the floor manager

────────────────────────────────────
  The first 20 minutes

  Target …someone at the table within 3 minutes

  Watch the two holes
    the first 40 min after open
    the rotation changeover  ⇒ stagger it

  Match the group: two to a four-top
  Cover first …large groups ／ name requests ／ returning

  Offer the second when the glass is a THIRD left,
  and name the drink: "the same again?"
────────────────────────────────────

Common objections

"You can't measure service like this"

⇒ You are not measuring service. You are measuring a wait, which is a fact, and it predicts the bill.

"We're too busy to time tables"

⇒ Ten tables a night, one number each. Two weeks. Then stop.

"Some groups just don't spend"

⚠️ True. That is why you compare bills against wait time rather than looking at bills alone. The pattern shows through the noise.

"Staggering the rotation is unfair on the girls"

⇒ Ask them. Most prefer a floor that never empties, because an empty floor is where they sit unpaid.

"Our staff know to offer a second drink"

⇒ Then the second-to-first ratio will be healthy. Check it before assuming.

This week

One   time ten tables a night for two weeks: seated to first entertainer
Two   write the bill next to each one
Third look specifically at tables seated during the changeover
Four  stagger the rotation so half change fifteen minutes later
Five  move the second-drink offer to a third remaining, and name the drink

Doing it by hand, or on a screen

Timing tables by hand for two weeks is the right way to start. Keeping it is the part that fails.

By hand …a notebook at the floor manager's station, then tallying against bills
On screen …seating time and the bill already on the same record, so the wait
           can be read against what the table spent without a second list

⇒ tasteck records seating and the bill against the same booking, so wait time and spend sit together rather than in a notebook and a till roll.

⚠️ Seating time has to be entered when it happens. Entered at the end of the night from memory, it will be neat, wrong, and worse than not having it.

Wrapping up

• The bill is largely settled before the second drink
• Measure seating to first entertainer — under three minutes, and the table settles
• Long waits cluster at opening and at the rotation changeover. Stagger the changeover
• Match the group size; two to a four-top beats one
• Offer the second round at a third remaining, and name the drink
• Fix the structure, not the effort. "Try harder" makes it worse

A quiet table is usually not a quiet customer. It is a table that decided, in the first twenty minutes, that this was a one-drink kind of place.