Thursday, July 15, 2010

Dueling Stats

Last month’s Arbitron-Research Director revelation that 22.2% of PPM cume is “P-1” and delivers 52.4% of the average station’s total average quarter hours (Ratio: 2.36:1) spotlighted how radically the numbers programmers have learned to use to track performance have changed.

Diary averages remain 35.7% of the average station’s cume is “P-1,” delivering 70.4% of average quarter hours. (Ratio: 1.97:1)

Does the difference in the ratios mean that having one of your heaviest users in the PPM sample is 20% more important than it has been to achieve consistent performance in diary samples?

Other new metrics which also require further investigation:

Diary turnover ratio, your total cume divided by average quarter hour persons audience, equals the number of times a song or announcement must be played before about half your cume will hear it two times in a week.

With a diary review or making use of software like Maximiser or PD Advantage you can calculate how many days per week your average diarykeeper writes down in their diary.

4.1 days per week is the best I've seen. 3.6 days per week is a little on the low side. That additional half day of average "regularity" can make a MAJOR difference in your diary-based ratings.

In PPM, since meters are in motion, on average, almost two thirds of the hours in a 24 hour day and most of them are in use seven days a week, the need to extrapolate such things using math goes away. Using meters it’s possible to see exactly how many days per week, times per day and average minutes per occasion the average panelist listens to radio overall and also to your station specifically. For example, DMR has stated that 94% of PPM panelists consistently use the same amount of radio week after week, with the heaviest users turning on the radio 31 times per week for an average of 15 hours. Sixteen of those occasions are spent with their favorite station.

Diary studies for years have shown that you get most of your quarter hours from "at work" listening and most of the average station's cume comes from "in car." A typical station that had 60 25-54 10:00 am to 7:00 pm weekday diaries that reported "in car" listening to a station, 14 "at home" and 15 "at work." Those 15 "at work" diaries reported 346 quarter hours of total listening, while the 60 "car" books contributed 378 quarter hours. Typically, it can take four times the cume (diaries) from "in car" to equal the same number of quarter hours from typical "at work" listeners. People think that cume builds average quarter hour. In fact, IN CAR listening builds cume and AT WORK builds average quarter hour - and it is equally important to have BOTH. A healthy station could have at least 70% of its cume "in car" and a similar percentage of its quarter hour "at work."

PPM replicates that, making it clear that in car is where most people switch stations a lot and at work they tend to dial around a lot less.

What does that say about traditional turnover/reach and frequency (OES)-based music rotations?

PPM also appears to find more males who listen at work which diaries have failed to register, but diaries report more female at work listening than PPM does.

What's real?

It’s going to be interesting to learn more about the study by Alan Burns and Associates released today which reportedly shows one-quarter of all women who listen to adult contemporary and CHR formats believe there’s no station in their market that understands them. “And that leads to erosion in radio usage,” Burns says.

Interviews with more than 2,000 female radio listeners found that two-thirds of women who think no station “gets” them are listening to radio less as a result.
“Either their interests and values have changed, and radio hasn’t kept up, or radio has changed in ways they don’t appreciate,” Burns adds.

So which is it? PPM accurately measuring real drops in women’s usage of radio? Or, a bias built into diary vs PPM methodology?

If getting more men who listen at work into our targets is a key to doing better in PPM ratings, will radio move further away from the (long term) tastes of females in pursuit of (short term) higher shares?


Anonymous said...

Was there an average age of the women not satisfied with what radio is offering? Harmonious

Jerry Del Colliano said...

Arbitron VP of Programming Services Gary Marince, according to Inside Radio, told the Jacobs Media Summer School at the Conclave in Minneapolis that the average radio listening occasion lasts ten minutes.

And, the most frequently occurring duration is only two minutes.

Marince was quoted as warning “against over-promoting an upcoming feature or teasing the wrong features”.

Why do you think listeners spend so little time listening?

One reason is the technology that Arbitron employs to generate PPM numbers. They encode radio signals. Then build portable devices that respondents carry around on their bodies. These devices pick up any encoded signal it “hears” and thus reports it as true listening.

So, if you’re 75 years old and love oldies, but you walk past an encoded signal wearing a meter that picks up an r&b/hip-hop station, then you are Ludacris listener whether you really are or are not.

Putting that little slight of hand aside, the real message is that radio is simply catching up to consumers in general who have shorter attention spans and want what they want when they want it.