Audience measurement quantifies the people in an audience and their characteristics, in relation to the media outlets and devices they use to consume media content and advertising. It encompasses television viewership, radio listenership, newspaper and magazine readership, and the consumption of online content. And it is primarily used by media owners, publishers and advertisers to assess the performance of media content and set advertising rates.
TV viewership is typically measured by means of a device hooked to the TV, called the people meter. Resembling a little box, the device reports which programme is being watched when the TV is switched on. It is accompanied with a remote control with push buttons allotted to each member of the household, so that individuals may identify themselves when they are viewing the TV.
The prime advantage of the people meter versus diary-based methods is that it is immune to forgetfulness. Diary keepers remember their favourite programmes, while lesser shows tend to be under-reported.
The key disadvantage of the people meter is that it involves the viewer’s proactive involvement, though to a much lesser extent than the diary method. Newer audience measurement methods are based on hypersonic facial recognition technologies. These solutions combine biometric authentication and neural network processing to passively detect and identify viewers.
Similar to the identification method used by smartphones like iPhone X, passive people meters identify individuals based on the spacing and contour of eyes, nose, ears, lips, chin and other facial features.
The incorporation of facial recognition technology into TVs started around 2012. These smart devices, which come with built-in camera and facial and voice recognition software, can recognize individuals, and turn on the channels and web applications that they like. The potential to apply this technology to track audiences more accurately and at lower costs, has existed ever since.
The ratings currency for measuring audio is the portable people meter (PPM). PPM panellists carry these pager-like devices that record the audio they listen to, throughout the day.
Developed by Arbitron (now Nielsen Audio), the PPM measures audiences exposed or listening to the wide array of media channels including radio and television. It detects hidden audio tones within a network’s audio stream, logging each time it finds such a signal. These masked audio tones are subliminally inserted into the audio feed via an encoder.
The latest PPM meters (PPM 360), introduced in 2010, use cellular telephone technology to transmit the data to the research agency’s servers.
Over the top (OTT) service providers such as Netflix, Amazon Video, YouTube and Hulu, that distribute programmes over the net, obtain audience information through internet protocols (TCP/IP). When users connect with the media server, details such as IP address and device information, location, browser details and so on are saved and maintained in the server’s log. This information may be used to track audiences. (Note: Messaging and voice calling services such as WhatsApp, WeChat and Skype are also OTT services).
IP address based information has the drawbacks discussed in the section Web Analytics, of Chapter Digital Marketing. For greater accuracy, OTT services require their customers to log-in when they connect with their service.
Internet Protocol TV (IPTV) also streams content to customers via internet protocol. The difference however, is OTT content is streamed over the public internet. IPTV services, on the other hand, use fully-controlled private, dedicated networks which cost more but can result in better quality of service.
IPTV services may be classified as live, time-shifted or catch-up (show replays) and video on demand (VOD) where user may choose from a catalogue of videos.
Like cable TV, IPTV usually works with set-top box (STB) connected to the user’s TV. It benefits from tighter control and guaranteed bandwidth, but, since the STB is tied to one network and one device, unlike OTT, it cannot serve content to viewers on the go.
Where there is a set-top box, audience viewership information is collected by means of an application installed in the box.
Information obtained via a return path from a set-top box or from user’s device as applicable for OTT platforms, is referred to as return path data (RPD).
Its sheer size and granularity gives RPD an edge over methods that rely on audience samples, especially as media consumption has become so fragmented. Within the context of its service platform, RPD yields greater accuracies, and enables a deeper understanding of audience behaviour. It facilitates ‘long tail reporting’, i.e., tiny audiences can be accurately measured and reported. Reports can also be generated at a more granular level on any of the dimensions — time, geography, demography.
Beyond audience measurement, the applications extend into content marketing, acquisition and development; subscriber marketing, acquisition and retention; loyalty schemes and selling advertising space.
Where consumers’ purchasing data is accessible, RPD can yield clearer insights on the short-term sales impact of advertising campaigns.
What is apparent however, is the great diversity of devices and platforms, yielding audience information that differs in structure and content, and is owned by multiple players. Combining all this information to form a single currency that cuts across platforms and across screens and devices, is a task that audience measurement service providers have been grappling with for quite some time.
Note: To find content on MarketingMind type the acronym ‘MM’ followed by your query into the search bar. For example, if you enter ‘mm consumer analytics’ into Chrome’s search bar, relevant pages from MarketingMind will appear in Google’s result pages.