It’s radio ratings time again. Here’s why the numbers are rarely simple

The JNLR can seem like ‘fake news’, but the underlying data is vital to stations – still

Pat Kenny’s Newstalk show is 30 minutes longer than it used to be - which should help him add listeners in the JNLR radio ratings. Photograph: Frank Miller/The Irish Times.
Pat Kenny’s Newstalk show is 30 minutes longer than it used to be - which should help him add listeners in the JNLR radio ratings. Photograph: Frank Miller/The Irish Times.

Media is a very straightforward business, a now retired broadcasting executive once told me: it is about building an audience, as big an audience as possible, and then selling that audience to advertisers. It was amazing to him how many companies couldn’t seem to do it.

For many media businesses – those with a mix of revenue streams to manage – it’s a little more complex than this. But commercial radio is still one form of media where pretty much 100 per cent of the revenue comes from advertising and sponsorship. Radio should really be as simple as selling audiences to advertisers.

This is the point in the presentation where the next slide that flashes up is an image of a tailors’ tape and the reminder that you cannot sell an audience if you cannot measure it.

Measurement is everything, which is why every quarter, there are a rash of headlines about how many listeners radio presenters and stations have won or lost. Later this week, it will be “JNLR” time again.

READ MORE

The Joint National Listenership Research (JNLR) survey is a weird mix of celebrity story, in which sad or happy faces can be attributed to Ireland's best-paid broadcasters, and a more prosaic, data-based report that nevertheless cuts to the heart of the businesses that pay those broadcasters' mortgages.

Alas, some JNLR “books” contain more genuine drama than others. Indeed, the most eye-catching takes have often been based on misinterpretations of the numbers, which is a polite way of saying they are fake news.

Mistakes

The most common misinterpretations – let’s call them mistakes – occur in the wake of changes to a radio station’s schedule. Changes to the schedule really mess up a station’s JNLR report, effectively making the numbers meaningless for a time.

One single, dull fact about the research methodology is critical for understanding the data: the JNLR may be published quarterly, but it is a rolling year-long survey.

Take, for example, the performance of Newstalk Breakfast, the "challenger" show to the long-dominant radio institution that is RTÉ Radio 1's Morning Ireland.

In the last JNLR survey, published in February, the figure for Newstalk Breakfast was 122,000. In February 2016, when the presenting team in situ was still Ivan Yates and Chris Donoghue, it was 171,000, while in July 2016, the survey published in the month Yates left the weekday show, it was 163,000.

In a post-FM world, radio audiences will be measured in real time, making it easier for stations to identify what is and isn't working, as well as helping them pinpoint who is listening to what.

But it is not the case that the new presenter line-up, in place since September, had “lost” 41,000 of the old team’s listeners or that the show had 49,000 fewer listeners than it did a year earlier.

The audience for a particular time slot on a particular radio station is derived from questionnaire answers sought by Ipsos MRBI researchers over a 12-month period. The total sample for the year exceeds 16,000 people.

So the 122,000 number relates to the answers of everybody who was questioned about their radio habits over the course of the calendar year 2016. Some of those people, at the time they were surveyed, were listening to Yates and Donoghue, some of them were listening to summer fill-ins and some of them were listening to Paul Williams, Shane Coleman and Colette Fitzpatrick (who has since left Newstalk).

The next JNLR survey (of people interviewed between April 2016 and the end of March 2017), and the next one after that, may well hint at the direction the audience is going, but a "true" number for the post-Yates Newstalk Breakfast won't be known until next October.

And even then it won’t be comparable to the previous audience because the “old” show had a three-hour slot, while the new show is only two hours long. Yes, nuance is the altar on which JNLR stories go to die.

Although there have been occasions when misleadingly partial information was passed along to journalists (who do not have independent access to the data), inter-station rivalry helps keep everybody more or less honest. On this point, let’s leave behind the perception that “everyone’s a winner” on JNLR day. While there have been times in the past when overall growth in listeners for demographic reasons was a tide that lifted all radio boats, in recent surveys it has not been hard to uncover losers.

In a post-FM world, radio audiences will be measured in real time, making it easier for stations to identify what is and isn’t working, as well as helping them pinpoint who is listening to what.

That’s some way off. In the meantime, the independent radio sector has ditched exclamation-heavy JNLR press releases in favour of jointly promoting the overall strength of radio audiences to advertisers.

This co-operation is a response to a deepening woe: Irish radio has joined television and news media in losing millions to Google and Facebook. Competition from other audio sources doesn't help. But the most pressing conundrum for radio is not that audiences are switching off – advertisers are.