Last Orders, Lock-Ins and the Hour You Cannot Bill

Every UK venue has an hour that is fully staffed, fully licensed, and produces almost nothing.

It runs from last orders to the moment the room is clear. The tills are shut, the cellar is done, the SIA cover is still on the clock, and the only thing being produced is drinking-up time.

Most operators treat it as a fixed cost of the licence. It is not — it is a schedule you chose, and it is one of the few costs at close you can actually move.

What the hour actually contains

Break it into three parts, because they behave differently.

Drinking-up time. Whatever your premises licence and local conditions require. Not negotiable, and not the part worth attacking.

Clearance. Getting people from "finishing a drink" to "out the door." Length depends almost entirely on how the last hour before last orders was run, not on how loudly anyone shouts.

Close-down. Tills cashed, float counted, cellar locked, glass collected. This happens regardless of when the room empties, and it is the part most likely to be running longer than anyone has measured.

The mistake is treating all three as one block called "close." Only one of them is fixed by the licence.

The number nobody has

How long, on average, between last orders and the room being clear — by night of week.

Almost no venue can answer this, because nobody writes the second timestamp down. The first one is on the clock; the second one is when someone happens to look up.

Two columns on the close-down sheet fix it permanently:

Last orders called   __:__
Room clear           __:__

Four weeks of that gives you a distribution rather than a feeling, and the distribution is what makes the conversation possible with your door team.

Why clearance is set an hour earlier

The single largest driver of a slow clear-out is what happened between the last hour and last orders.

A room that ran a heavy final hour — big rounds landing at the bar right up to the bell — clears slowly, because everyone has a full drink and the drinking-up clock starts from the last one served.

Serving hard right up to last orders buys you revenue in one hour and spends it in the next, in door supervision and bar staff you are still paying.

That is not an argument for cutting service early. It is an argument for knowing the exchange rate. If your last twenty minutes reliably produce a certain amount of trade and reliably add a certain number of minutes to clearance, you can price the decision instead of arguing about it.

Where the SIA cover sits

Door supervision is usually rostered to a fixed finish, and the fixed finish is usually set from the licence rather than from measurement.

Two things worth checking:

Whether the finish matches your actual clear time, or whether you are routinely paying for thirty minutes after the room is empty — or worse, running short at the moment you most need cover.

Whether the same finish is right on every night. A Tuesday that clears in fifteen minutes and a Saturday that takes fifty are being staffed identically in a lot of rooms.

The cost of getting this wrong is asymmetric. Over-staffing costs money. Under-staffing at chucking-out time costs an incident, and an incident costs a licensing conversation.

⚠️ Do not cut door cover on the basis of an average. Cut it on the basis of your worst night in the last eight weeks, and only if that night still holds.

The lock-in question

Worth stating plainly: a lock-in where drink is being supplied outside your permitted hours is a licensing problem, not a revenue strategy, and the person who carries it is your DPS.

There are lawful ways to keep people in the building — a private function, a genuinely non-licensable arrangement — and they all depend on your specific premises licence and local conditions.

That is a question for your licensing solicitor, not for an operations article, and not for a decision made at 01:40 by whoever is on.

What belongs in operations is the rule being written down and known by the closing manager, so the decision is not being made under pressure by someone guessing.

The close-down half

The part that is genuinely yours to move.

Time your close-down once. Not an estimate — actually time it, on a normal night. Most venues find one or two steps take far longer than anyone assumed, and the usual suspects are glass collection and the float count.

Glass collection that starts at close is a close-down cost. Glass collection that runs through the last hour is absorbed into service and costs nothing extra.

The float count is frequently done once, at the end, by one person, under fatigue. Splitting it — counting the tills that shut early as they shut — moves the work into time you were already paying for.

Neither of these is clever. Both routinely remove real minutes from the paid tail.

Three numbers to hold

Last orders to room clear, by night of week. The distribution, not the average. Your door roster should match the upper end, not the middle.

Close-down duration, timed once per quarter. It drifts upward and nobody notices until someone times it.

Trade in the final twenty minutes before last orders. Against the clearance minutes it adds. This is the exchange rate, and most rooms have never calculated it.

Where the record has to sit

The two timestamps are the whole thing, and they need to sit next to the shift they belong to — otherwise you have a column of times and no way to join them to who was on, how busy it was, or what the door cover cost that night.

tasteck keeps shift records, staffing, and the night's trade together, so how long does Saturday take to clear, and what are we paying for that is one view rather than a clipboard nobody transcribes.

The hour after last orders is going to happen. Knowing its length is what turns it from a licensing cost into a rostering decision.