How to cut waiting times

The four waits in a service, and which one hurts

A customer waits four times, and the four do not weigh the same. The first and the last are the ones people remember, because during those nothing is happening: they are looking around for somebody.

Waiting for the dish, by contrast, is tolerated well as long as there is a drink on the table and somebody has said how long it will be. Which is why shortening the kitchen is usually the most expensive fix and the least noticed.

Sitting down and waiting to be taken     5–10 min   ← hurts
Waiting for the first dish               8–15 min     tolerated
Between courses                          5–10 min     tolerated
Asking for the bill and it arriving      5–15 min   ← hurts
Paying and getting the change back       3–10 min   ← hurts a lot

The three stretches that hurt add up to 13 to 35 minutes
on a table that only spent about 90 actually eating.

The first wait: nobody should have to hunt for a waiter

A customer who has just sat down reads the menu for two minutes and then looks up. That moment decides whether the service will feel attentive or careless, and it happens before you have done anything at all.

What works is not moving faster, it is the order not depending on somebody walking past: taking it on the waiter's phone without going back to the bar, or letting the guest order themselves when they are ready.

• Let the waiter take the order without walking to the bar and back.
• Let whoever already knows what they want order without waiting.
• Drinks first, always: a table with a drink waits twice as long without complaining.
• If you are swamped, say so when you seat them. A flagged wait counts as half.

The last wait: payment is where turnover disappears

A table that has finished and waits twenty minutes between asking for the bill, it arriving, paying and getting change back is an occupied table earning nothing. In a two-hour service that is 17 % of your capacity thrown away.

And it multiplies when four people are paying separately: four trips with the card machine, four confirmations and an argument about who ordered the second bottle of wine.

It is the stretch with the most to gain, because the customer has already decided to leave: shortening it takes nothing away from them.

The dead time between courses

If the table waits twenty minutes between starter and main, it is almost never because the kitchen is slow: it is that nobody fired the next course. The call comes from the floor, which can see the table, and if the floor is swamped the course sits there with nobody knowing.

A kitchen screen shared between floor and kitchen solves half the problem, because both sides see the same thing instead of guessing.

What does not work, even though it looks like it should

Three common reflexes that make service worse rather than fixing it.

• Adding waiters at peak time without changing the flow: more people making the same trips.
• Shortening the meal by rushing people. Turn the table once and lose the customer for good.
• Speeding up the kitchen when the problem is on the floor: dishes come out and go cold on the pass.
• Changing the menu so it cooks faster without having measured where the time went.

How to find your own bottleneck

Before touching anything, measure one service by hand. A notebook and four columns: time seated, time ordered, time of first dish, time they left. Ten tables is enough to see the pattern.

It almost always comes out the same — the two ends — but it is worth checking in your own venue before investing in solving a problem you do not have.

Frequently asked questions