Attendance and related metrics (such as proportion of tickets sold etc) are valuable measures of an event’s popularity.
It is important to recognise that the event population extends beyond spectators or audience members to include participants (e.g. athletes, performers or delegates), venue staff, officials, media personnel, VIP guests and volunteers. You should include these groups in your attendance figures and, if possible, segment the overall event population along these lines to enable deeper analysis of impacts associated with each group.
Accurate attendance data is also essential to the wider range of impact assessments that rely on scaling up sample-based survey results to reflect the behaviour of the overall event population. If you don’t know the size of that population, you can’t extrapolate your survey data with any degree of certainty.
What to measure
We recommend you prioritise two key indicators when calculating event attendance:
- Total attendance: the cumulative number of people who attend your event across all days, sessions, venues and admission types (i.e. paid and free)
- Unique attendees: the number of different people who attend (the economic calculator asks for data on ‘spectators’ and ‘other attendees’)
Event attendance: How to measure it
For single-day ticketed events, measuring attendance can be relatively straightforward.
For multi-day, multi-site or non-ticketed events, however, it can be far more challenging. To get the right total, you may need to collate returns from multiple ticket sellers, draw on entry data from a range of venues or rely on crowd counts taken at different times in different places. A further level of complexity is added by the need to measure unique attendance, which requires you to be able to identify people who came along more than once to avoid double counting of these attendees.
At ticketed events, you should always consider the following when calculating attendance:
- Your primary sources will be box office data, ticket sales and ticket distribution
- The number of tickets distributed or sold does not always equal the attendance at the event (i.e. some ticket purchasers may not turn up)
- Buyers of tickets may not be the people who use them
- Re-use policies may see a single ticket used by more than one person if it is re-sold should the original holder leave early
At non-ticketed events, the OECD recommends the use of the following crowd-counting techniques:
- Video analysis is most suitable for static linear events (single-session events at which attendees stand in the same place along a linear route)
- Crowd density analysis is best for static non-linear events (single-session events at which the entire crowd is present at the same time)
- Tally counters and/or video analysis can be used for dynamic linear events (those at which attendees move along a linear route)
- A combination of all three are needed for dynamic non-linear events (multi-session and/or multi-day events)
For large-scale events it may also be possible to estimate attendance numbers retrospectively by surveying residents of the host community to assess their engagement with it.
Event attendance calculation in action
Case study: 2021 UN Climate Change Conference (COP26)