Skip to content
Empty padded rows of seats in a darkened cinema auditorium with a lit balcony - nobody is shown.

Cinemas and halls - counted seats against sold seats

Tickets sold and people in the auditorium are two figures. Most venues only have the first.

The box office system knows how many seats were sold for the 8 pm show. How many people actually sat in the auditorium, it doesn't know. Usually the two figures are close, and nobody asks. Where they drift apart - through complimentary tickets nobody recorded, guests who switch screens, bookings that never turn up - nobody notices, because nobody counts.

This article is about reconciling the two figures: in cinemas, theatres, concert halls and anywhere people hold a seat for a particular performance. How the reconciliation works mathematically is covered in Occupancy is a balance, not a count. Here the question is what it does for running a venue.

Why the two figures drift apart

A deviation isn't automatically a problem. But each direction means something different:

DirectionTypical causesWhat it means
More people than ticketscomplimentary tickets without a booking, guests from the screen next door, gaps in entry controllost revenue and, in a full auditorium, a safety question
Fewer people than ticketsbooked seats that never show, group bookings with drop-outsseats that could have been sold, and a concession stand planned for too many
Both figures equala show running as plannedthe normal case - and the proof of it

The first row is the one people think of first. The second is often more valuable for planning: a venue that knows ten per cent of booked seats regularly stay empty on Tuesday evenings plans programme, staff and concessions differently.

Reconciling each show

The reconciliation itself is simple, and that's what makes it useful. For each auditorium and show, these sit side by side:

  • the expected value - the seats sold or registered,
  • the measured headcount - from counting at the auditorium's entrances,
  • the deviation, absolute and in per cent,
  • the capacity used in the auditorium.

If the deviation exceeds a threshold - 10 per cent by default - it counts as a discrepancy and is recorded. One condition is easy to overlook in a cinema: counting has to start cleanly between two shows. Carry the 6 pm audience into the figure for the 8 pm show and you're comparing nothing.

The history shows where checking pays off

A single deviation says little. A premiere with press seats, a school screening with accompanying adults - there are good reasons why more people sit in the auditorium one evening than tickets were sold.

It gets revealing over months. When discrepancies cluster in particular screens, days or times, it's no longer chance but a pattern. Entry checks can then be placed where the history shows deviations, rather than putting the same number of staff on every door.

That's the operational lever: not checking every show, but checking the right ones.

Picture a venue where someone tears tickets at every auditorium entrance today. If each screen's occupancy is continuously reconciled with tickets sold and the count starts afresh after every show, someone only needs to check where the figures drift apart. The staff on the door are freed up for the concession stand or maintenance.

If you know where deviations arise, you no longer need checks everywhere.

Capacity use as a planning figure

Alongside the reconciliation, the same count supplies a second figure that programme planning lacks: how full a screen actually was, not how full it was sold.

  • Programme. Which film or play belongs in which auditorium? Capacity used per show over weeks shows where a smaller room would have done and where a bigger one was missing.
  • Concessions and staffing. The rush at the counter follows the people in the building, not the tickets sold. Planning intervals and admission times against actual occupancy is more precise. How footfall becomes a staffing plan is covered in The visitors who don't buy.
  • Cleaning. An auditorium that held 300 people needs something different after the show from one that held 40.

When a count triggers money

In some venues the headcount is more than a planning figure: it's the basis of a levy, a licence fee or a settlement between two businesses. Then it has to be not just right but able to withstand an audit.

So any figure money depends on needs a measurement that can be repeated, a history nobody alters after the fact, and a report that comes from the same source as the dashboard. How such a chain of evidence is built is covered in Capacity as a permit condition.

Beyond the cinema

The same principle applies wherever people hold a place for a set time: in theatres and concert halls, in lecture halls with registration, in conference halls with sign-up. For sponsors and partners of an event, how many registered people actually sat in the room often matters more than the registration count itself. For meeting rooms, where a booking takes the place of a ticket, the relationship between booking and presence is covered in Occupied versus paid-for space.

What belongs in the tender

  1. Can an expected value be reconciled against the measured headcount per auditorium and show?
  2. Is the deviation shown in absolute terms and in per cent, and can the threshold be set per room?
  3. How does counting start afresh between two shows?
  4. How long is the history of deviations kept, and can it be analysed by room, day and time?
  5. Who is notified of a deviation, and when?
  6. If the figure is the basis of a settlement: how is it evidenced, and who can check it?

How we do it

Capacity verification in the ANALYSIT Counting System reconciles the live count against the entered expected value for each auditorium and show: actual, expected, delta in absolute terms and per cent, and capacity used, plus a session timeline for today and a “Next Up” countdown to the next show. Expected value and threshold can be adjusted per show, the count refreshes every 15 seconds, and the peak is sampled every minute.

A discrepancy is flagged as soon as the percentage deviation exceeds the site's threshold - 10 per cent by default. When a session ends, the deviation alert is evaluated and delivered by email, SMS, WhatsApp, Slack or Teams, and the discrepancy history is available for 13 months. The counting data itself is stored on a one-minute grid for 730 days and can be exported as PDF, XLSX and CSV.

If you want to know how many people really sat in your auditorium, talk to us. The first month already shows the first evenings on which the figures drift apart.