What Your First Hour Is Really Costing

Ask why you open at the hour you open and the answer is usually historical. It was the hour when the venue opened before you, or the hour on the licence, or the hour that felt right when the room was a different business.

Very few venues have tested it, and the ones that have are frequently surprised in both directions.

What the hour costs

The full cost is larger than the payroll line, and the parts people forget are the ones that make the decision.

Staff on the clock. Usually two to four people. This is the number everyone thinks of.

Prep that happens anyway. Some of it would happen regardless — the setup, the ice, the till. Opening earlier does not add this cost; it moves it. Do not count it twice.

Energy and utilities. Small per hour, real over a year.

The tail cost. An earlier open frequently means the same staff are still there at close, which pushes into overtime or a second shift. This is the one that gets missed, and in some rooms it is the largest single component.

The opportunity cost of the roster. A person rostered for the quiet hour is a person not rostered for the busy one, if headcount is capped.

What the hour earns

Two things, and they are usually confused with each other.

Direct revenue. What the guests in that hour spend. Easy to measure, usually small.

Anchoring. Guests who arrive in the quiet hour and stay through the busy one. Their whole tab is not first-hour revenue, but it exists because you were open when they wanted to start.

That second one is why comparing first-hour revenue to first-hour labour cost is the wrong test — it undercounts. It is also where venues talk themselves into keeping an hour that is genuinely not earning, because "those people stay all night" is easy to assert and rarely measured.

The measurable version: for guests whose first drink lands in the opening hour, what is their total tab across the night, and how many of them are there.

That is a single query if tabs are attached to guests and timestamps are kept. It answers the anchoring question directly rather than by argument.

The test that actually works

Do not model it. Change it for a fixed period and measure.

Pick one night of week. Move the open by one hour. Run it for four weeks.

One night, because changing all of them at once means you cannot attribute anything. Four weeks, because two is not enough to survive a rainy Thursday and a public holiday.

What to compare, against the same night in the four weeks prior and against a control night you did not change:

• Total revenue for the night
• Labour hours for the night
• Arrival curve — did the crowd shift or shrink
• Revenue per guest

The arrival curve is the most informative. If you open an hour later and the same guests arrive an hour later with the same spend, the hour was costing you and producing nothing. If the crowd shrinks and does not come back later, the hour was anchoring more than it appeared.

The result that surprises people

Two outcomes come up repeatedly, and they point in opposite directions.

The quiet hour was pure cost. The guests who used to arrive in it simply arrive later, spend the same, and the labour saving drops straight through. More common in rooms that inherited their hours.

The quiet hour was carrying the night. A small number of early guests were high-spend regulars who came because it was quiet, and they did not come back at the busy hour — they went elsewhere. Losing them cost multiples of the labour saved.

⚠️ You cannot tell which one you have without running it, and the floor's intuition splits roughly evenly. The staff working the quiet hour usually believe it is dead. The guests in it are frequently the ones who matter most.

The version that is cheaper to test

If moving the open feels too risky, test the staffing instead.

Keep the hour, cut the staffing in it to the minimum that can serve the volume, and see whether anything breaks. That isolates labour cost from anchoring value, and it is reversible on a single night.

Most rooms discover they were rostering the quiet hour for a crowd that stopped arriving in it some years ago.

The last hour, briefly

The same analysis applies at close and the answer is usually different.

The last hour tends to have worse revenue per guest and worse labour efficiency, but it carries something the first hour does not: the guests in it are frequently finishing a long tab, and closing early truncates the spend of people who were already going to spend.

Test it the same way. But expect the closing hour to defend itself more strongly than the opening one, and expect incident and staffing costs at close to be higher.

Three numbers before you decide

First-hour labour cost, fully loaded — including any tail into overtime.

Total night spend of guests whose first drink was in that hour — the anchoring number, and the one most rooms have never calculated.

Arrival curve for that night of week — so you can see whether the crowd moves or disappears when you change the hour.

Where the data has to come from

The anchoring number is the one that requires the join. It needs the first drink's timestamp attached to a guest, and the guest attached to the full tab across the night.

Without that, the question collapses back into first-hour revenue versus first-hour cost, which systematically under-values the hour and pushes toward closing it.

tasteck keeps tabs against guests with timestamps, so what do the people who arrive first actually spend by the end of the night is a number rather than a belief.

The hour is either earning or it is not. Four weeks on one night tells you which.