Why this number matters more than most

Table turnover is the operational figure that converts most directly into revenue. You do not need more marketing, more covers walking through the door, or a bigger room — you need the room you already have to serve more people in the same evening.

It is also the number most often "improved" in exactly the way that costs you money: by rushing the meal, which is the one part of the cycle the guest is actually paying for.

The distinction that makes this work is simple. Almost all the recoverable time is before the meal and after it. Neither is time a guest wants to spend.

Measure before you change anything

Two rules, and most restaurants break both.

Measure per table, not as a floor average. An average tells you nothing you can act on. What you want is that table 14 by the kitchen door turns once a night while table 6 turns three times — that is a specific problem with a specific cause.

Measure in four segments, not one. A single "occupancy time" number cannot tell you where to act.

# Segment Typical Yours to fix?
1 Seated → first order placed 4–12 min Yes — the biggest leak
2 Order placed → food delivered 12–25 min Partly — kitchen sequencing
3 Last course → bill requested 8–20 min Mostly the guest's
4 Bill requested → table reseated 8–20 min Yes — the second biggest

Run this for a fortnight, on every table, on real trading nights. A clipboard works. If your orders are recorded in a system, the order timestamp and the payment timestamp give you two of the four boundaries without any extra effort.

Most owners are wrong about which segment is their problem. That is the point of measuring first.

What the time is actually worth

A 20-table restaurant, two sittings a night, average order value ₹850, six days a week.

Recovering ten minutes per table:

  • 20 tables × 2 sittings = 40 turns a night
  • 10 min × 40 = 400 minutes of table time recovered
  • At 70-minute average occupancy, that is ≈ 5.7 additional covers-worth of table time
  • ≈ ₹4,845 a night
  • ≈ ₹1,26,000 a month, or ₹15 lakh a year

That is more than most independent restaurants make from marketing, and it comes entirely from time nobody wanted to spend.

Ten minutes is not ambitious. Four minutes off the wait to order and six off the reseat gap will usually do it.

Segment 1 — the wait to order

The largest and cheapest win.

On a full floor with two captains covering sixteen tables, five to ten minutes before anyone takes an order is normal. It delays the kitchen, delays the turn, and quietly kills second rounds — because a second round only happens if someone comes back to ask.

What works:

  • Named section ownership every shift. Shared responsibility means the table nobody owns waits longest.
  • A greeting within ninety seconds, even when the order will take longer. Guests tolerate waiting much better once they have been seen.
  • Menus on the table before the guest sits, not fetched afterwards.
  • Removing the dependency entirely. Letting guests order from their own phone takes this segment off the critical path — the order no longer waits for a person to be free. Whether that suits your room is a real question, answered honestly in QR code ordering: how it works and when it doesn't.

Segment 2 — order to food

Partly the kitchen's capacity, but a surprising amount is sequencing rather than speed.

If tickets are cooked in the order they were clipped to a rail rather than the order they arrived, you have manufactured wait time that no additional cook removes. An early table waits behind a late one because of where a piece of paper landed.

The fixes, cheapest first: a strict oldest-first rule on the rail that one named person owns; prep times recorded against dishes so mixed tickets can be sequenced; or moving the queue to a screen where oldest-first is automatic. The wider treatment is in restaurant operations.

Do not attack this segment by cutting portion prep quality. You will win four minutes and lose the reason people came.

Segment 3 — leave it alone, mostly

This is the guest lingering after the last course. It is the segment owners most want to compress and the one they should touch least.

The legitimate version is making it easy to ask for the bill — being visible, being reachable, not disappearing for eleven minutes. The illegitimate version is clearing plates early, hovering, or dimming the lights. That works exactly once per guest, and the return visit you lose is worth more than the sitting you gained. The arithmetic on that trade is in how to increase repeat customers.

One genuinely useful change: if your bill takes eleven minutes to produce because the order is being re-keyed at a till, that delay is yours, not the guest's. Fix it there.

Segment 4 — the reseat gap

The invisible one, and usually the second biggest.

A table pays and leaves at 8:20. It is wiped at 8:24. At 8:31 a party of four is told there is a twenty-minute wait and leaves. The table was free for eleven minutes and nobody with a waiting guest knew.

This fails because the floor plan lives in someone's head and heads go out of date every few minutes during a rush.

What fixes it:

  • A shared, current view of the floor rather than one person's mental model — the detail is in table management.
  • A "cleaning" state. Without it a vacated table is either invisible to the host or visible and dirty.
  • A clearing routine. Whose job is it to reset a table, and how quickly? Usually nobody's, which is why it takes eleven minutes.
  • Knowing which table turns next, which also lets you quote honest wait times at the door.

Turnover is not enough on its own

Track occupancy alongside it — the share of seats actually filled across a service.

A restaurant seating parties of two at six-tops all evening can post excellent turnover figures and still leave a third of its seats empty. Turnover counts parties; occupancy counts people, and people are what pay the rent.

The seating discipline that fixes this — smallest table that fits the party, large tables held during peak with a cut-off — is covered in table management.

Common mistakes

Compressing segment 3. The one segment guests notice, and the one that costs repeat visits.

Treating a floor average as actionable. It hides the tables and segments that matter.

Adding staff before fixing sequencing. More people running a badly ordered queue is a more expensive badly ordered queue.

Ignoring the reseat gap because it happens after the guest has gone and nobody is watching.

Optimising turnover while ignoring occupancy. Improving a number that is not the one paying your rent.

Changing three things at once. Then not knowing which one worked.

What to do this fortnight

  1. Record four timestamps per table for two weeks. Clipboard is fine.
  2. Calculate each segment's average and its worst tables. Per table, not per floor.
  3. Pick the single worst segment. It will probably be 1 or 4.
  4. Change one thing and measure the same fortnight again.
  5. Add a cleaning state to however you track the floor, whatever else you do.
  6. Track occupancy alongside turnover so you are not optimising the wrong number.

The short version

  • Measure per table, in four segments, for a fortnight. Most owners are wrong about which segment is their problem.
  • Recoverable time is segments 1 and 4 — the wait to order and the reseat gap. Neither is time a guest wants.
  • Ten minutes per table across 40 turns is roughly ₹1.26 lakh a month.
  • Leave segment 3 alone. Rushing the meal buys one sitting and costs a returning guest.
  • Add a cleaning state. It is the cheapest fix for the reseat gap.
  • Track occupancy alongside turnover, or you will optimise parties instead of people.