Your Pour Cost Is a Schedule Problem

Pour cost is the number every bar owner knows and almost nobody can move.

You take inventory, you calculate it, it comes in three points above where it should be, and the response is some combination of: buy jiggers, retrain the staff, count more often, change the well.

Sometimes that works. More often the number comes back the same next month, and the conclusion is that this is just what a busy bar costs.

Before accepting that, split the number by shift. In most rooms the variance is not spread evenly across the week — it is concentrated, and it is concentrated on people.

Why the monthly number hides everything

A monthly pour cost is one number covering roughly thirty nights, four to eight bartenders, and two completely different kinds of business.

Your Tuesday is a different bar from your Saturday. Different drinks, different pace, different staff, different guests. Averaging them produces a figure that describes no night that actually happened.

The number you want is pour cost by shift block — or as close as your counting cadence allows.

Weekly counts on the well and the fast-moving bottles get you most of the way there. You do not need a full inventory to see a pattern; you need the same twelve bottles counted at the same point in the week, consistently, for six weeks.

Consistency of method beats completeness of scope. A partial count done identically every week tells you more than a full count done differently each month.

What the split usually shows

Three patterns come up repeatedly.

Pattern one — one shift is three to five points above the others. Same drinks, same specs, same guests. This is the most common finding and it is almost never theft. It is usually free-pouring under pressure, or a bartender who over-builds because they were trained somewhere with different specs.

Pattern two — variance rises with volume, but faster than volume. Your busiest nights should run slightly higher pour cost. Speed costs accuracy, and that is a real trade. But if a night that does 40% more revenue runs eight points worse, you have crossed from acceptable speed-cost into a staffing problem — the bar is under-manned and everyone is compensating by pouring faster and measuring less.

Pattern three — one product is the entire variance. Everything else is in range and one bottle is not. This is rarely a people problem. It is a spec problem, a portioning problem, or a promotion nobody costed.

The three patterns have three different fixes and only one of them is training.

The staffing version of the problem

Pattern two is the one worth dwelling on because it is the one that gets misdiagnosed most.

When a bar is short one person on a busy night, the bartenders do not slow down. They speed up, and the first thing to go is measurement. Nobody decides to over-pour — it happens in the gap between the guest waiting and the jigger being one extra motion.

That means a chunk of your pour cost is being set on the roster, days before anyone touches a bottle.

The check: plot pour cost against drinks-per-bartender-hour for each shift over six weeks. If the line bends upward past a certain point, you have found your bar's actual capacity — the pace beyond which accuracy collapses.

That number is more useful than any pour-cost target, because it is actionable on Monday when you build the schedule.

Adding one bartender to a shift costs a known amount. If it recovers four points of pour cost on your highest-volume night, the arithmetic is usually not close.

The bottles that are worth counting

Full inventory is expensive in labour and most venues do it too infrequently for the numbers to be useful.

The practical alternative is a short list, counted often:

• The four to six well spirits
• Any bottle in your top ten by revenue
• Anything currently on promotion
• Anything a staff member has flagged

That is usually ten to fifteen bottles and takes fifteen minutes. Done weekly at the same point in the week, it produces a trend line. Done monthly with a full count, it produces a number with no context.

A trend on twelve bottles beats a snapshot on two hundred.

The promotion nobody costed

Pattern three deserves its own check because it is the fastest to fix and the easiest to miss.

Run the variance by product for one month. If a single item is carrying the number, look at what changed around it: a new spec, a new garnish, a bigger glass, a two-for-one that ran longer than intended, a bottle that got moved to a harder-to-reach spot and is now being poured by feel.

Glassware is a genuinely common cause. A change of glass with the same free-pour count is a silent price change, and it shows up in exactly one product's variance.

What to do about a shift that runs hot

If one shift is consistently above the others and the volume explanation does not hold, the conversation is worth having — but the framing matters.

Coming in with "your pour cost is bad" produces defensiveness and no information. Coming in with the actual number, alongside the other shifts, and asking what is different about that night, usually produces the answer in about ninety seconds. Staff know. They are frequently relieved to be asked.

The answers are mundane: the well is laid out awkwardly for a left-handed pour, the ice bin position forces a reach, a specific regular orders something built off-spec every week, the jiggers were in the dishwasher.

None of those are visible from a P&L, and all of them are cheap to fix once named.

What a healthy number actually is

Ranges vary by concept and by market, so treat any published benchmark as a starting point rather than a target.

What matters more than the level is the spread. A bar running 22% across every shift is in better shape than one averaging 20% with shifts ranging from 15% to 29%. The first has a cost. The second has a control problem, and the average is hiding it.

Reducing the spread is usually easier than reducing the level, and it is where the recoverable money is.

The three things to track

Pour cost by shift, weekly, on the short bottle list. Six weeks of this tells you more than a year of monthly full counts.

Drinks per bartender hour, by shift. Your capacity curve. The point where pour cost bends is your real staffing threshold.

Variance by product, monthly. Catches spec and promotion problems before they become a quarter.

Where the numbers have to come together

The obstacle in most venues is not the counting. It is that the count lives in a spreadsheet, the sales live in the POS, and the roster lives somewhere else entirely — so nobody ever puts pour cost and staffing on the same page.

That join is the whole insight. Pour cost by itself is a scolding. Pour cost next to who was working and how many drinks they made is a schedule you can change.

tasteck holds shifts, sales, and staff on the same records, so the question which shifts run hot, and were they short-handed is one query rather than three exports and an afternoon.

The fix for pour cost is frequently on the roster. You just cannot see it while the two numbers live in different files.