Why Did Breakfast Spend Go Up? Price Creep vs. Volume

Food cost went up. A classic hotel mystery. Was it more guests, more eggs, more coffee — or did supplier prices quietly creep up while nobody was looking?

It is one of the most common questions a GM or controller asks, and one of the hardest to answer honestly, because the answer is buried across dozens of supplier PDFs that nobody has time to line up side by side.

Two different problems wearing the same costume.

More volume usually means more demand. More guests ate breakfast, so the hotel bought more bagels. That is a good problem, and the correct response to it is nothing at all.

A higher unit price is a different animal. That is supplier price drift: margin erosion that goes unnoticed because no single invoice looks alarming. Nobody sends a letter announcing an increase on plain bagels. It simply appears on the next delivery, in the same place on the page, in the same font as always.

The two problems call for opposite reactions, which is exactly why guessing between them is expensive.

What breakfast actually looked like at one hotel.

We are piloting Atrium with a hotel in Niagara Falls. Real invoices, real breakfast, eight months of ongoing use. Over that stretch Atrium flagged 24 product price increases at the property.

The shape of those 24 is the part worth sitting with:

• They averaged 12.3%, with a median of 11.1%.
• The gentlest was 5.1% — small enough that questioning it on the page would feel unreasonable.
• The sharpest was 31.9%: house blend coffee, from $160.50 to $211.69.

Same coffee, same supplier, same delivery day. The only thing that moved was the number in the price column.

Twenty-four separate items over eight months, at one hotel. Not one dramatic event — twenty-four quiet ones, each arriving on an ordinary delivery day and each too small on its own to argue with.

Volume is the other half of the question.

A price increase is only half an answer. The same line carries a quantity, and the quantity is what tells you whether the hotel bought more bagels or simply paid more for the same delivery.

Atrium keeps both. Unit price against its own history, quantity against its own history, on the same line — so the follow-up gets narrower than the original question. Which of these do we accept, and which do we call about?

Why nobody catches it in the moment.

Invoices arrive to be paid, not to be studied. The person approving them is checking that the delivery happened and that the total looks plausible. Against the invoice total, a single line moving 5.1% is a rounding error on the page. It is not supposed to register.

Then it repeats. The new price quietly becomes the price, and by the time anybody looks back there is nothing left to compare it against except memory. Twelve weeks later the increase is simply what things cost.

That is the whole mechanism. Not fraud, not incompetence — just the fact that no human holds four months of unit prices in their head while approving a Tuesday delivery.

Why the P&L cannot tell you which one happened.

The monthly food cost line is a single number, and a single number is a sum. It cannot distinguish between serving more breakfasts and paying more for each one. Both push it up. Only one of them should make you happy.

Occupancy explains some of it. Occupancy explains none of a 31.9% move on a coffee item that did not change. Separating the two is not a reporting nicety — it decides whether the right response is to celebrate or to call the rep.

A category total tells you what you paid. Only the line item tells you why.

How to separate them without opening a single PDF.

The check itself is mechanical. Take each recurring item, line it up against its own history, and ask two questions: did the unit price move, and did the quantity move? Everything then sorts into one of four boxes.

• Same price, more units — volume. Demand went up. Nothing to do.
• Higher price, same units — price drift, and worth a conversation.
• Higher price, more units — both at once, with the price problem now amplified by the volume.
• Higher price, fewer units — the quiet one. Spend looks flat while the rate climbs underneath it.

That fourth box survives the longest, because the total never moves. Spend looks steady. The rate does not. One more reason the P&L is the wrong place to go looking.

What the hotel does with the answer.

At that property, the estimated annual impact of those 24 increases is $4,613. Alongside them, Atrium surfaced 24 comparable-supplier opportunities worth about $796 a month, with pack sizes normalized so a six-pack case is compared against a six-pack case rather than against a single unit.

Neither figure is a promise. They are places worth looking. The hotel may well decide the coffee is worth the increase because guests notice it, and push back on a line no guest will ever taste. That is exactly the right use of this — Atrium narrows the question, the team answers it.

The question behind the question.

When a GM asks why breakfast spend went up, they are rarely asking for a number. They are asking whether something is wrong. A total cannot answer that. A handful of lines carrying their own price history can, in about the time it takes to read them.

This is the visibility GMs, controllers, and purchasing teams should have without digging through PDFs — the heart of hotel invoice intelligence. Ten recent invoices are enough to see whether your breakfast lines have been moving.