Everything About Measurement Technology: The Top 5 Ways To Measure OOH Audiences

Just as in the world of online advertising, measuring audiences is critical in digital out-of-home advertising; a range of measurement technologies have been created to facilitate this.

Accurate measurement techniques allow media buyers to carefully optimize campaigns and to achieve better ROI by finding their target audience more easily. The same measuring instruments also allow media owners to accurately sell ad space and prove network value. 

Measurement technology is improving the customer experience by providing levels of personalization previously only seen online. Using smart technologies like video analytics software as part of programmatic advertising, advertisers can trigger content changes depending on who is looking at a digital display. 

In this article, we run through the five main ways you can measure your audience online and achieve these levels of personalized targeting; we also give our opinion on which is the most effective measurement technology. 

Infrared Array Sensors

Infrared array sensors work by placing the sensor above areas where digital screens are placed, or at building entrances, and they then detect rises in temperature to assess the number of people in that space and their positions. This thermal data image will be fed through software to provide information about footfall and the estimated number of people who might have seen a display.

Typically costing less than $200, this is a cost-effective audience measurement technique and can provide accurate data on impressions and dwell time (how long someone stands within a close proximity to a digital display). 

Where this measuring instrument is less effective, however, is in providing data that shows how many views a digital display has really received. Infrared array sensors pick up bodies within the vicinity, but there’s no way to know whether they were really engaging with the content.  

Mobile WiFi Tracking/ Location Data

Through data from mobile phones, GPS systems and WiFi-enabled cars, users can assess the size of an audience passing through a location. WiFI tracking works by tapping into a device’s automatic probe request for WiFi. All routers and WiFi devices receive a signal from any WiFi-enabled device and this is pinned to a precise location. This measurement instrument therefore gives insights about audience size and can help publishers prove inventory value to media buyers.

However, while this audience measuring instrument is relatively cost-effective (at $40-$60 per tag), it is also limited. It casts a wide net when it comes to counting people in a building or area and there is no way to truly determine the number of people who are seeing an ad display, only those who were likely to based on their location.  

There are also a lot of concerns around the invasive nature of WiFi tracking. Most often users do not opt-in or give permission to these types of measurement techniques which makes them morally questionable. 

Bluetooth Beacons

Bluetooth beacons work similarly to WiFi tracking. When signals are automatically sent out by a Bluetooth-enabled device, they are picked up by devices in a range of 20m to 300m and precise location can be tracked through the communication. 

For quantities of over 50-100, beacons cost around $10-15 per one. However, just like WiFi tracking, the data provided is limited. It cannot indicate exactly how many people are engaging with a digital display and no granular audience analysis data is provided.

3D Video Analytics - Sensor-Based Measurement

3D video analytics work by using machine learning and dual camera lenses (placed on ceilings) to detect people through depth perception. As accurate as the human eye, they allow users to count the traffic around a digital display through a camera and to assess how many people are really likely to have seen it.

Based on height, the software can distinguish the rough age of viewers and by tracking how long people stay in a certain area, it can provide data on dwell time.

However, whilst it offers high levels of accuracy, it is still limited in many ways. Data that can be gleaned is accurate, but does not indicate specific demographic insights or viewing behavior data in the way that AI video analytics can do.  

AI Video Analytics Technology - Camera-Based Measurement

AI video analytics technology uses cameras and computer-vision powered software (such as that provided by Raydiant) to provide extended audience insights.By using cameras, AI video analytics is not only able to offer accurate measurement data surrounding the number of bodies within the field of view of a display, but it can also supply granular insights on audience demographics and their viewing behaviors. This information can be collected using a combination of technologies; anonymous facial detection and body detection and goes beyond other measuring instruments. AI video analytics technology can even measure an audience member’s reaction to a digital display — whether they look happy or neutral, for example.Like other measurement technologies, AI video analytics can also indicate how long individuals spend lingering close to displayed content (dwell time). However, what sets it apart from the rest is that it is also able to distinguish between Likelihood To See and Opportunity To See, illustrating how many people were in the range of the display v.s. how many people actually engaged with it.  AI video analytics technology can also detect whether someone is wearing a face mask — data that has proved invaluable in recent years. 

The Best Method Of Measurement Technology:

With a wide range of analysis features it goes without saying that AI video analytics software is the most well-rounded and conclusive form of measurement technology out of those listed above. It offers high levels of accuracy and extended audience analysis and it provides incredible ROI and allows media buyers and owners to reach new heights of success. Best of all, it is also completely privacy compliant. Unlike video analysis that is facial recognition based, no identifiable information is ever collected or stored. Making it the perfect solution for industry wide applications. It can help businesses obtain the data needed to measure the effectiveness of on-premise networks and digital campaigns while still making the privacy of the public a top priority. Raydiant's audience analysis technology is built with a privacy first mindset and has a face blur  privacy feature built in. To learn more about our audience analysis software, click here.