Why your GP looks fine on paper but feels wrong in practice

Most operators know this feeling. The GP figure in the system looks reasonable. The accountant has not raised any alarm. The weekly report shows margins broadly in line with expectations. And yet the business never quite seems to have as much left over as the numbers suggest it should.

This is one of the most common and most underdiagnosed problems in hospitality finance. The GP figure is not accurate, but because it is plausible, nobody challenges it. The operation runs against a margin it has never actually achieved.

Why a GP figure can look right and be wrong

There are several ways this happens, and most of them are structural rather than anyone's fault.

The calculation uses invoices instead of actual consumption. The most common cause. If COGS is calculated by adding up what was spent with suppliers during a period, rather than from a proper stock count, the figure will be wrong whenever stock levels change. Build stock up and your GP looks artificially high. Run stock down and it looks artificially low. The average might look reasonable over a year, but week to week the number is not measuring what it claims to measure.

The stock count is consistently optimistic. When counts are rushed, sections get estimated. The people doing the count know roughly what is there and put in a number that looks right. Consistently optimistic estimates produce a consistently flattering GP. The variance between estimated and actual closing stock never shows up as a problem because the estimate is never tested against a careful count.

Unrecorded consumption is not captured. Staff meals, wastage from prep, the bottle that broke, the keg that was tapped and discarded because something was off. In a well-run operation these are tracked and accounted for. In most operations they disappear. The product was consumed, the cost is real, but the GP calculation does not know about it.

Delivery shortfalls are not credited. A supplier delivers eleven cases when the invoice says twelve. If the short delivery is noticed but no credit note is raised, the business paid for twelve and received eleven. That gap is invisible in the GP calculation. Over a year of weekly deliveries, these accumulate.

Recipe costs are out of date. If dishes were costed when ingredients were cheaper and the recipes have not been updated, the theoretical GP the operation is working towards is higher than what is achievable with current costs. The operation targets 68% and achieves 63%, but because the target was wrong to begin with, nobody knows to investigate the gap.

The danger of a plausible number

There is a specific risk in a GP figure that is wrong but not obviously wrong. A GP of 22% would prompt immediate concern. A GP of 64% when actual performance is 60% will not.

The operator makes decisions based on 64%. Prices are set on the assumption of a 64% margin. Staffing levels are justified against it. Investment decisions are made against it. The four-point gap between the assumed margin and the real one is not a rounding error. On a business turning £500,000 in F&B revenue per year, four points is £20,000 of margin that the business thought it had but did not.

The insidious part is that the operation adapts to the gap rather than closing it. GMs learn that the month-end always feels tighter than the weekly numbers suggested. Finance teams build in a mental buffer. The discrepancy becomes a known unknown that nobody officially names or resolves.

Understanding why most operators do not know their true GP helps clarify which of these structural problems applies to your operation and where the fix needs to start.

How to find out if your GP figure is real

The test is simple in principle: compare your GP calculated from invoices with your GP calculated from actual stock counts.

Take an opening stock count at the beginning of a period. Add everything purchased in. Take a closing stock count at the end. Calculate what was actually consumed. Divide by revenue. That is your actual COGS and your actual GP%.

If the figure from this calculation matches what your system shows, your data is probably reliable. If it is materially different, you have found the gap.

Most operators who do this exercise for the first time discover a discrepancy. Sometimes it is reassuring: the real GP is slightly better than assumed because stock has been building. More often, it is in the other direction: the real GP is lower, and the difference represents a specific problem that can now be investigated.

That investigation is where the value is. A 4-point GP gap has causes. It might be portioning drift. It might be a delivery reconciliation process that is not catching shortfalls. It might be unrecorded waste in a specific section. It might be a supplier who has been incrementally increasing invoice amounts and nobody has noticed. Each cause has a remedy. Without the comparison, the gap stays invisible.

The role of variance

The comparison between expected and actual GP produces a variance figure. This is not a number to fear. It is a number to understand.

Supplier shortfall: product that was invoiced but not received. Raise a credit note.

Recipe drift: the kitchen is using more than the recipe specifies. A training or portioning conversation.

Unrecorded consumption: product is disappearing without being accounted for. A process or security conversation.

Legitimate variation: a busy weekend, a high-spend table, a catering event. Nothing to worry about.

The variance tells you which conversation to have. Without it, all you have is a feeling that something is not quite right, and no place to start.

What changes when the data is right

Operators who move from estimated GP to verified GP describe the same experience. The first few weeks are sometimes uncomfortable: the real number is lower than the assumed one, and there are specific causes to investigate. After that, something changes.

The GP figure becomes a tool rather than a report. When it moves, you know why. When it holds, you know what is working. Decisions about pricing, suppliers, portions, and purchasing are made against real information rather than a plausible approximation.

The business does not automatically become more profitable. But the information needed to make it more profitable is finally visible.

That is the shift. Not a dramatic transformation. Just the difference between running on data you trust and running on data you have quietly learned not to.

Hops connects stock takes, POS sales, and purchase invoices in one place, so the GP figure you see is calculated from what actually happened. For operators who have been working with an estimated margin, the first month is almost always illuminating.