Modelled traffic counts, mobile location data and panel surveys are the three established ways to measure out-of-home. Each has a credibility problem in Australia.
Alternative approaches
Modelled traffic counts are easy to attack. Mobile location data is legally uncomfortable post-2022. Panel surveys are slow and expensive. Each has credibility problems that advertisers, and their lawyers, are increasingly unwilling to sign off.
How anonymous counting works
A camera above a digital screen runs an edge model. The model counts people who looked at the screen for more than two seconds, estimates their age/gender bracket, and discards the frame. The event is an aggregate: 18–34 male, 4.2s dwell, 10am–11am. No individual is identified. No data leaves Australia.
Why the number is defensible
Advertisers brief against a measured count: how many people looked at the screen, for how long, in which bracket. No individual is identified, and no data leaves Australia.
For publishers the trade is hardware: a camera above each screen, an edge unit in the cabinet, and a weekly audience report. In return, modelled traffic counts are replaced by a measured figure.
Questions to ask a publisher
- How is the audience counted?
- Is any individual identified?
- Where is the data hosted?
- What is the sample size per screen per week?
Ask for the sample size in writing. If a publisher cannot supply one, the number is not defensible.