We came up with the estimate after Nielsen released a summary of listening in the 51 markets they survey.
The graphs to the left compare Arbitron PPM, Arbitron diary, and Nielsen sticker diary listening estimates for three key measures: reach (cume), TSL (time spent listening), and AQH (average quarter-hour) rating.
PPM cume estimates are minimally higher than in either Arbitron’s or Nielsen’s diary, but considerably lower in more important measures.
For example, PPM estimates that TSL in PPM markets is a little under 12 hours. In markets still measured by diary, TSL is about seventeen and a half hours, almost 50% higher. Nielsen using their own sticker diary estimates TSL at nearly 23 hours a week, nearly twice Arbitron’s PPM TSL.
Arbitron admits that PPM estimates are lower, which is why it created the misguided 70 is the new 100 sales campaign aimed at media buyers. Their goal was to convince media buyers that they could buy only 70 rating points in a PPM market with the confidence that they were really getting 100 rating points.
We called it Arbitron’s Fuzzy Math back in September when we first wrote about it. Asking media buyers to pay more for less struck us as a little optimistic and maybe even naive.
No official word from Arbitron on how the campaign is going, but word from our clients in PPM markets is that media buyers don’t buy it. Lower PPM ratings means media buyers demand lower rates.
It stands to reason. Media buyers are not going to pay more for less. In a soft economy, most media buyers are going to try to pay less for more.A radio station switching from diary ratings to PPM who walks in with an AQH rating 30% lower than before is going to get beat up.
Either the buyer will pay less for the same number of points, or get 30% more spots than she got before. In either case, PPM has cost the station, either in cash or units.
What would happen if it turned out that Arbitron’s PPM estimates were wrong?What if radio was actually a 16.5 AQH rating medium that Nielsen says it is rather than the 9.1 AQH rating medium that Arbitron says it is?
Media buyers would have to pay higher rates, or get fewer spots. In either case, radio wins.
Radio wins financially, but equally importantly, it wins psychologically. Lower PPM numbers portray a medium in trouble. The uninformed and those who have a financial interest in perpetuating the myth that radio is in decline can point to Arbitron’s lower PPM estimates and use them to beat radio up.
In 2004 the Radio Advertising Bureau commissioned a study to examine the financial impact of PPM. The study concluded that PPM would create a positive revenue swing of $433 million.
PPM apologists will argue that PPM would have had a positive impact on revenue, but that the recession wiped out the gains. One could as easily make the argument that the transition to PPM exacerbated the impact of the recession and worsened the declines.
What is irrefutable is that since the RAB’s upbeat report on the positive impact of PPM, radio revenue has dropped nearly $6 billion, about 30%.
As we pointed out in our first post on the subject, PPM was already rolling out well before the recession hit. If PPM were going to bring new dollars to radio, we should have seen it then, before revenues started declining across the country. Instead, early PPM markets were already in a tail-spin before the recession.
In any case, it is interesting to do a “back of the envelope” estimate of what radio might be billing if PPM markets were instead using Nielsen estimates to sell time.
We estimate that in 2009 the top 50 markets billed about $9 billion. If radio could sell using the 16.5 Nielsen rating rather than Arbitron’s 9 rating, radio could generate up to an additional $7 billion at the same cost per point.
It’s just a guess, but think about the implications. Radio’s down about $6 billion. Where would radio be today if we were a billion dollars ahead instead of $6 billion behind.
Makes you wonder, doesn’t it?
Note: Arbitron does not release national estimates comparable to the Nielsen estimates, so to create the Arbitron numbers we averaged five PPM markets and five diary markets. We've found that market estimates are very similar across markets (in technical terms there is very little dispersion), so a larger sample of markets would not change the results significantly. We hope that one day Arbitron releases estimates similar to Nielsen's so we can fine-tune our numbers. If they do, we'll report them here.