• -
3d render of neon lights - live

Going Live – Seize the Multichannel Present, not the Analogue Past

Category : Blogs

Some types of analysis take longer than others. But, it’s generally felt these provide deeper insight, or more value than quicker ones. However, there are some types of analysis that are take longer, but provide less value.

I was reminded recently of this point in conversation with a colleague about the model by which TV and Radio advertising ratings are calculated.

I say reminded because it’s about 15 years since I was involved in the print industry body’s attempts in the UK to update its methodologies in light of the changes in customer behaviour from digital channels. At the time, digital disruption in the newspaper industry was revolutionising how consumers read newspapers and magazines. This shift suddenly meant that the print industry’s method for calculating readership (print circulation of each title, multiplied by a “readership” factor for the number of times the average copy would be read by different people) no longer reflected the reality of consumption; lots of people read newspapers online, which didn’t show up in traditional readership figures.

To be fair, the newspaper industry reacted pretty quickly to this change (or threat) to their business model, but other media methods have remained stubbornly resistant to change.

TV and Radio, for instance, both have panel systems, whereby everything that those on the panel watch or listen to is recorded or diarised, and then multiplied by a “weighting” factor to work out how many other people like them would also have watched the same programmes.

So, you probably already know that The Great British Bake Off final, in which Nadiya Hussein was crowded Bake Off Champion, was the most watched programme in British television in 2015. Or more recently, you may well know that one of this year’s peak TV programmes so far was watching the Olympic cycling at 11.1 million viewers. Or you may know that Radio 1 is in trouble this year because Nick Grimshaw’s breakfast radio show has lost about 1 million listeners in the first half of the year.

And we know this because TV and Radio use panels to determine their viewership and listenership figures respectively, using broadly similar panel-based methodologies.

And how many people are on those panels?

Given the nuances of the population, and the plethora of new channels that consumers could be watching or listening to, you would have thought that this would be fairly large. But actually, it’s still only 5000 people on each of the panels.

5000 people – the same size of panel when there were only 4 or 5 TV channels and many fewer radio stations. So the measurement model is still the same as that in the 1970s and 80s.

And that measurement is what advertisers pay for; it’s what advertising rates are based on for both TV and Radio. And as a result, it is generally regarded to favour the larger broadcasters; the nuances of niche channels simply don’t show up, or are too small, to be correctly measured in the official ratings for TV and for radio.

However, digital insight managers are able to deliver much more rapidly. They have the capability to analyse digital consumers across delivery channels (digital player, app, web site), which means that digital insight managers have a far greater level of detail about their audiences than these panels could deliver. That’s because they can operate in real time, rather than having to wait for the results of diary surveys, which in the case of radio can take up to 3 months – yes, 3 months – to come in.

To give you a specific example, the last set of Radio audience (RAJAR) figures came out in early August – I write this in mid-October. The next set of figures are due out at the end of October for Q3 (July-September) 2016. In a world where customers can be targeted by retailers in real time, and companies like Netflix can give you tailored recommendations as soon as you have finished watching or listening to your last programme, it still takes 6-7 weeks, after the end of the relevant quarter, to process the data of 5000 people. It truly is an analogue method in a digital age. The only reason why it is still just about relevant is most people listen to radio using analogue equipment, rather than digital devices.

In the meantime (that is, since the last release of data), digital insight teams will have been able to be identify, create and target particular segments for, by example, listeners for the Reading Festival or the Proms; indeed, it’s quite possible to have built and used mini segments within the Proms – those who tuned into Baroque performances, or Brahms, or the CBeebies Prom, or the David Bowie one. All of these would have been extremely relevant for this period, and could help to understand cross-over listening recommendations but will not feature in the RAJAR data. Digital managers for radio news channels will no doubt have been able to identify those who voted for Brexit, and those that didn’t, simply from their listening patterns, and will probably now be repeating the process for Trump vs Clinton in the US.

RAJAR’s Q3 figures will cover the entire period of UK radio consumption of the Olympics, the Paralympics, the Labour Party leadership election, the Last Night of the Proms, the change of UK Prime Minister, political conference season, and almost all of the post-Brexit news fallout. In a news-driven world, it seems incredible that almost all of this is “old-news” already, and yet, Radio in particular focuses its official figures on the past, rather than the present, which its digital teams can deliver insights very rapidly.

At Station10, we use a method called insight sprints as a way we deliver more valuable insight more quickly for clients; if you are interested in how this might help speed up your insight response times, please get in touch.

Sign up for our newsletter