Audience measurement faces tricky questions

Back from Athens, Julie MacManus reports on the 2018 ASI International Radio & Audio and Television & Video Conferences

It’s been a while since Socrates advocated the practice of thoughtful questioning to examine ideas, so he was probably looking down proudly on the ASI in Athens last week at the large, throwable (yet not always catchable), microphone tossed between conference delegates.

And there are some tricky questions facing the TV and radio audience measurement community as rapid change becomes the new normal. Fundamentally, the industry needs to reach consensus on what TV and Radio actually is before it is able to measure it.

While the consumer may not care, or even be aware, as platforms begin to blur the distinction, whether they are watching linear TV, Broadcaster VOD, YouTube or Netflix, the consolidation of audience metrics still has some way to go. Especially when some services don’t want to be measured. While two of the FAANGs, Facebook and Google were at the ASI calling for greater collaboration, Kantar and GfK both told the conference that their lawyers had issued advice that it is legally inappropriate to publish SVOD programme level data since it would violate the T&Cs users sign up to. While, over in the US, Nielsen are publishing figures, and are completely confident in their right to do so.

Wouldn’t the ASI be the perfect forum for these research companies to reach a consensus on what they can do? Apparently not, since the legal eagles say this would constitute collusion.

So why the sensitivity? Interestingly, the Nielsen viewing figures show how Netflix Originals (on which spend is projected to be $12-13B this year) perform well at launch but tail off quickly, while older library content accounts for the most viewing and retains audiences steadily. [advert position=”left”]

For media owners, like Disney, this raises questions about whether they are benefiting enough from making their content available via these services. And it’s not just Hollywood studios facing these questions; the BBC World Service found that just 7% of users was able to identify them as the source of their news stories consumed via Facebook.

Increased fragmentation also requires a disproportionate amount of spend on audience measurement compared to their levels of usage. As a ballpark, 15-20% of BARB’s operational costs goes on measuring the 1.3% of additional viewing which devices, other than the TV set, provide.

And that still doesn’t include the 47 minutes a day of viewing on a TV set which isn’t BARB reported (presumably largely SVOD and gaming). This unidentified viewing has grown by 58% for all audiences since 2015 and accounts for almost 40% of young adults’ viewing.

If you take the line, suggested by a recent Deloitte research programme, that the current behaviour of young people will be replicated by all ages between 5-8 years later, then this needs tackling since an industry currency which isn’t comprehensive will struggle to be trusted. And trust and transparency in audience metrics is crucial in a world where the FBI have launched formal investigation into media buying practices.

Finally, we were reminded how accurate media measurement which genuinely reflects the whole population has an importance which extends beyond trading deals. 5 years ago, when South Africa’s Broadcast Research Council first came to the ASI in Vienna, their media currencies were a hangover from the days of apartheid, heavily skewed toward a non-black, metropolitan population.

With learnings from the conference and the help of the ASI community, they have since been able to produce TV and radio currencies which reflect the many languages and experiences of the whole of the new South Africa, forming a baseline for the media which will serve them. Nicely done, ASI.

Julie MacManus is Head of Mediatel Connected