Summary

This event report looks at one possible solution to the problems researchers face in measuring audiences across multiple platforms.
Read more

The proposal is to shift from a focus on specific platforms - TV, print, radio etc - to one that embraces the ways in which people consume content, so audiences would be measured in terms of viewing moving pictures, listening and reading.
While there is a certain logic to this, the main parties involved - advertisers, media owners and agencies - all have different priorities in what they want from audience measurement.
There are also issues around the way content is distributed and the devices used to access it.

Watching, listening and reading: A new classification for better audience measurement

Emily Barley
Warc

"We need a house of new definitions," according to Leendert van Meerem, of LVMR Media and Research Consultants. For example, "we're not talking about television anymore – it's video," he told an ESOMAR multi-platform media seminar held in Amsterdam in June 2015.

The past was simple, with just television, radio, print and outdoor and established measurement tools for each of these – combining metering and survey data. But advances in technology – especially the internet and mobile – have changed all that. "The big confusion is the word digital", said van Meerem. And "only analogue people – old people – feel that digital is anything new". For everyone else it is their 'normal'.

Most consumers are just getting on with using the services available. "The consumer is not in chaos at all. He is using SMS and WhatsApp side by side, and he's completely used to programmatic buying," he said.

But advertisers and researchers face a different challenge in updating their measurement approaches for a complicated and fragmented media landscape. Video is perhaps the best example of this – "the great media chaos: is YouTube television?" Nowadays instead of just TV and cinema, there is video from multiple sources, including online newsbrands, radio publishers and user-generated content.

Van Meerem suggested a new classification of media to change how content is considered and measured – focusing on viewing, listening and reading. He set out some of the options for measuring audiences in each of these areas.

Video: viewing moving pictures

Researchers could attempt to measure total viewing on all devices at any moment from all sources.
Or they could measure whatever they reliably can. Traditionally advertisers and researchers were reliant on the figures reported by TV ratings companies, some of which, including the German TV ratings company, AGF, are now incorporating YouTube data.
Another option is to measure all streaming media that is viewed above a certain threshold, thereby limiting measurement to the most popular platforms.

Audio: listening

Traditionally radio has been measured through survey data.
Meters for audio matching or coding (fingerprinting and watermarking) have been developed, but costs need to be reduced if these are to be rolled out.
"The problem is they give lower audio ratings," said van Meerem. And media owners are resistant to that.

Text: reading

"Digital reading is small, but it is growing," said van Meerem. And with the decline of newspapers the desire of publishers to get "their lost readers back" is dominant.
Audience measurement needs to reinvent reading from the online world and transform it into an 'average issue readership' that can be compared across platforms.
Advertisers want to know who is doing what, with which device. And though progress has been made, van Meerem said they're not there yet: "We can measure details that we do not want, and we cannot measure mainstream details that we do want."

No magic bullet

Even with this simplification there are still two factors that will add complexity. Firstly, the way content is transported or distributed to people, for example whether it is analogue or digital, push or pull.

Secondly, researchers want to know what kind of device people are using to access content. At the moment, this includes analogue devices, audio-only devices, fixed digital screens and mobile digital screens. "And don't think this is the end of it," said van Meerem. "We will have all sizes of all screens – hanging on the wall, lying on the table, sitting in your pocket."

"The internet is nowhere and everywhere", he continued, and it has no simple definition because nowadays "it is everything". This is incredibly difficult for researchers to work with, and "the only positive thing is that we can measure it".

But researchers also need to step outside of the data streams, van Meerem argued. "Sometimes we think it is push, but it is pull. We have to go away from the pure technical definitions, and look at it from the consumer point of view."

What is more, researchers know that different types of client organisations have different priorities in what they want from audience measurement:

Advertisers are driven by cost considerations, chiefly wanting lower advertising costs for the same quality by being smarter or more targeted with their media buy.
For media owners, measurement is "the oil that he needs to run", said van Meerem. They want an audience measurement system that favours the particular media they own, as this feeds through to their sales.
When it comes to media agencies, "they have the least problems with price fluctuations", said van Meerem. Their main concern is to have more campaigns.

Which means accurate audience measurement is not in everyone's best interest – and in many cases the people with the most data about a media channel (media owners) also potentially have the most to lose from reduced ratings.

And then these problems are further compounded, as van Meerem explained: "The problem is we are dependent on what these consumers are using as an operating system."

"We need measurement techniques that are independent of operating system," he continued. ""We need a way to follow the content."

But for now researchers need to take a step back and reconsider their priorities and client needs, not just what they can do. "Rough is enough… we are able now to measure in detail, second by second, but do we need it? A good overview is better than unreliable granularity," van Meerem concluded.

About the author

Emily Barley is Editorial Assistant at Warc.