Average cost per ticket is the metric every campaign report leads with, and it is almost always wrong in the same direction. It divides all spend by all sales, which quietly credits paid media with every ticket that would have sold anyway. On a show with strong organic demand, that can be most of them.

The question that matters

Nobody actually needs to know the average. The decision in front of a marketing team is always about the next increment: if we add ten thousand euros to this date this week, how many additional tickets appear? That is marginal cost per incremental ticket, and it behaves nothing like the average.

Early in a run with real demand, average cost looks excellent and marginal cost is high, because the campaign is standing in front of buyers who were already coming. Later, when the warm audience is exhausted, the average still looks acceptable while the marginal cost has climbed past the ticket price. Budgets get cut at exactly the wrong moment because the average never moved.

How to actually measure it

There are three practical approaches, in ascending order of rigour.

• Geo holdouts. Withhold spend in comparable markets and compare sell-through. Clean, slow, and only viable on multi-market runs.
• Budget steps with matched dates. Move spend up or down in defined steps on one date while a comparable date holds steady, then read the difference in pace. Faster, noisier, usable on most tours.
• Modelled incrementality against the sales curve. Fit the expected curve for a date from comparable runs, then attribute the deviation to the spend change. Requires ticket-level data and a reference set, and is the only approach that scales across a full calendar.

What the number tells you

Marginal cost per incremental ticket has one job: to be compared with the net value of a ticket. Above it, spend is destroying margin. Below it, spend is producing it, and the gap tells you how much room is left.

Read across a run, the curve of marginal cost over time is the clearest picture of a campaign there is. It usually falls in the first days as targeting resolves, sits flat through the productive middle, then rises steeply as the addressable audience thins. The moment it crosses ticket value is the moment to move budget to another date, not the moment the average finally looks bad.

The uncomfortable implication

Measured this way, some campaigns that report beautifully are producing very little. A show that would have sold out regardless will make any channel look efficient. This is why platform-reported return is not a safe basis for allocation on a strong run: the better the show, the more flattering the number, and the less the spend is doing.

It also means the honest answer is sometimes to stop. A date at ninety-five percent with three weeks left does not need media. It needs to be left alone while the budget goes to the Tuesday show in the next city that is sitting at forty.

Getting to that answer requires sold tickets per date per day next to spend per date per day. With those two series, marginal cost is arithmetic. Without them, it is a conversation.