Guaranteed numbers and actual attendance — who pays for the twelve who didn't show?
The client guaranteed a hundred. Eighty-eight walked in. The bill has already been paid by the hotel, the only question is whether any of it comes back.
The email usually arrives three weeks after the event, from someone you have never spoken to. "Hi — we've received an invoice for 100 delegates. We were 88. Could you correct this?"
And then begins an argument neither side is equipped to win, because the person writing the email wasn't in the room when the number was agreed, and the person who agreed it has left, is on leave, or remembers it differently. Eventually you credit the twelve. Not because you were wrong, but because having the argument costs more than giving in.
This post is about why that argument happens, what the credit note actually costs, and what it takes to stop it happening again.
What a guarantee actually is
A final guarantee is not an estimate. It is the point at which the client takes on the risk of their own attendance, and the point at which the hotel converts a number into money it cannot get back.
The mechanics are much the same everywhere. An inquiry arrives with a projected headcount, the contract carries an attrition schedule that allows the number to come down in steps — perhaps freely until 60 days out, ten percent until 30 days, nothing after the guarantee deadline — and the client then confirms a final number a set period before arrival. Seventy-two hours, three business days, five, depending on what you wrote. After that point the number can go up but not down.
That is where the asymmetry sits, and it is worth stating plainly: for you, the guarantee deadline is the moment the cost becomes irreversible. For the client, it is still an estimate. The kitchen has ordered. The staffing roster is set. The meeting room has come out of sale, and the room block was released too late for anyone else to take it. The client, on their side, has replied to an email.
The kitchen also rarely produces exactly to the guarantee. Most set for a few percent over, because the alternative is running out of the main course at noon with thirty people in the queue. That overproduction is a real cost that never reaches anyone's invoice at all, and that nobody measures.
The arithmetic, with actual numbers
Say the 24-hour delegate rate is NOK 2,450. Twelve people who didn't show is NOK 29,400. Over a two-night conference it is close to NOK 59,000. Not catastrophic for a hotel, but not loose change either — and it is rarely one event.
What matters is what those NOK 29,400 are made of, because that is where the conversation with the client should actually start. Roughly speaking, a delegate rate breaks down into accommodation, food and beverage, meeting space and technical. Those four behave completely differently when someone fails to turn up.
The food is gone. The raw materials were bought, prepared and in many cases thrown away. The labor is gone in the same way — the roster was built for a hundred, and nobody sends a banqueting server home at nine because twelve fewer arrived. The meeting room is also lost, but differently: the room was never a cost, it was capacity you could not sell to anyone else. Accommodation is the only line where there is a genuine chance the loss was contained, if the rooms went back into sale early enough for someone to actually buy them.
That distinction is the whole point, and it is why both "we invoice everything" and "we credit everything" are bad answers. A contract that treats food and labor more strictly than bedrooms is not only fairer. It is far easier to defend in an argument three weeks later, because the logic is one the client recognizes from running their own business.
The credit note costs more than the amount on it
When you credit the twelve to make the conversation go away, you are not buying peace. You are buying a precedent.
The client learns that the guarantee is not binding. Next year they guarantee a hundred again, arrive with eighty-five, and send the same email, this time citing the fact that you corrected it last time. Their event coordinator mentions it to a colleague at another company, because the industry is small and everyone talks. You may never see the pattern in your own numbers either, because the credit note lands in the accounts as an adjustment rather than as a written-off claim.
NOK 29,400 is not a one-off cost. It is an annual subscription to the same conversation.
Why you lose the argument even when you are right
Nearly always for the same reason: you cannot document what was confirmed, when, and by whom.
The guarantee typically arrives as a reply in an email thread on a Friday afternoon, often from someone other than the person who signed the contract, and often phrased as "we're expecting around 100". The salesperson reads it, nods, and goes home. The number is never recorded against the contract. The kitchen gets it on a printed sheet or over the phone. Housekeeping gets a different version. By the time the billing file is assembled two weeks later, nobody can say whether it was 100 or "around 100", or whether someone rang on the Monday and said 92.
Then comes the part that really hurts: the argument ends up being conducted between the client's finance team and your finance team. Two parties who were both absent when the agreement was made, working from a document that does not contain the answer.
We have written before about why it matters whether you own your data or are only renting access to it. This is the most down-to-earth version of that problem we know of. The history exists — it is just sitting in an inbox, a phone call, and the memory of someone who no longer works here.
Four things that actually work
Put the reduction in the contract rather than in the invoice. A clause allowing the client to reduce by up to ten percent until 72 hours before arrival costs you less than the argument you would otherwise have afterwards — and it has an effect most people do not expect: clients who know they have ten percent of room guarantee more accurately, not less. What disappears is the defensive under-guaranteeing. Without such a clause, experienced clients guarantee low and ask to add people later, which is the worst possible outcome for the kitchen.
Require the confirmation in writing, to one address, and record it the minute it arrives. Not in a thread, not to a salesperson's personal inbox. The number needs a timestamp, a named sender and a place in the contract — not a place in somebody's email.
Write the answer into the invoice before the objection arrives. An invoice line that says "Delegate rate, 100" invites the email. One that says "Guaranteed 100, attended 88, invoiced 100 per clause 4.2, confirmed by Kari Nordmann 11 Sept 14:22" does not. The difference in dispute volume is not marginal. Most objections are not disagreement — they are a request for an explanation, dressed up as a claim.
Measure your own accuracy. Few people do this, and it is probably the most valuable of the four. Track guaranteed numbers against actual attendance, by client, across a year. You will find the variance is not randomly distributed: some clients run fifteen percent over, every single time. That is not a billing problem, it is a pricing input. Either you price it in, or you have a conversation with the client in which you have data and they have a hunch. That conversation is much easier than the credit note one.
What this means for your system
Most of the above is contract work and sales culture. But one requirement is purely technical, and it is where most systems fall down: the guaranteed number has to be one number, in one place, with a history.
Every change needs a timestamp, a person and a source. The number has to reach the kitchen, housekeeping and the technical team without anyone retyping it somewhere else. And both the guarantee and the actual attendance have to travel all the way to the invoice line, so that the basis is visible to whoever has to answer the objection — without digging through last year's email.
When you are evaluating systems, this is a concrete thing to ask in the demo, and it is easy to check: ask them to show you where the guaranteed number is changed, who sees the change, and what ends up printed on the invoice line. Most can show you a field. Fewer can show you the history behind the field. We have written more about what the demo tends to leave out, but this is one of the faster tests we know.
The caveat
None of this solves the problem on its own. A system can make the guarantee traceable and the billing basis auditable, but it cannot enforce a guarantee that your own sales team will not enforce. If the culture is that large clients get a credit note regardless, the documentation just becomes a better archive of claims you gave up.
Nor should you expect to win the argument if your own delivery had holes in it. If you overbooked the meeting room, swapped a dish without telling anyone, or moved the group into a lesser space, then the twelve delegates are suddenly not the only topic on the table and nor should they be.
And for the record: on invoice flow into the accounting system, established vendors with long PMS heritage still have more mature integration than we do. We think we are stronger on event production itself — the function sheet, the covers, who actually got told about the change but we are not going to pretend we have come equally far on everything.
Who pays for the twelve who didn't show? You do, from the moment the guarantee deadline passed. The only open question is whether you get it back, and that is not decided in the invoice argument three weeks later. It is decided by what the contract said, and by whether you can show exactly when the number was confirmed and by whom.


