Pat Welsh
by Pat Welsh
There’s been a lot of information coming out about the PPM. Some of it revealing, some of it frightening, some of it expected and all of it preliminary. The truth is, we still don’t know exactly how much of what we’re seeing is the trend of the future with electronic measurement, how much may be unique to the first two markets, Philadelphia and Houston, or how much can be attributed to simple anomalies since we haven’t seen that many months of data.
Still, there are a few trends that seem to be developing. From a programming perspective, these are the things from the first nine months that are making the biggest impact:
There’s no more phantom cume – Cumes are much bigger with the PPM, an overwhelming trend that we’ve seen in every country where electronic measurement has been tested or implemented. We’ve seen so far that station cumes virtually double with the PPM automatically while TSL drops significantly.
The big controversy now is about “what’s more important, cume or TSL?” It depends on how you define important. On the one hand, cume plays a much bigger role in determining your ratings, but which one do you have more control over? What we’re about to find out is how much you can manipulate your TSL with “listen longer” tactics. Which contest, features, programs and personalities really keep people listening longer? How much of TSL is the amount of time that people have to listen and how much is attributable to compelling programming?
People Change Their Minds I – Possibly the biggest change with PPM is as of yet getting little attention: P-1 isn’t as clear-cut a concept as we thought. Since the diary methodology takes a quick snapshot of a listener’s alleged preferences, the concept of P-1 is pretty straight forward. But with the PPM tracking people over time, we’re seeing the gray areas that you would expect with some respondents’ P-1 station changes over time.
A person may listen to station A as their favorite for a month, then switch partisanship to station B the next. These “soft P-1’s” muddy the waters for the concepts of cume and P-1. This makes a lot of sense when you stop to think about it. But what we don’t know yet is what kind of P-1 churn we should expect. Does this happen more with certain demos, ethnic groups or format partisans? These are the kinds of benchmarks that need to be established before we can answer questions about whether a station is doing a good job or a bad one.
Arbitron has just released a study on this phenomenon, although it was based on tests conducted in 2006 in Houston before all stations were encoded. Still, the presentation, available here on the Arbitron Web site, is an interesting read.
People Change Their Minds II – Another foreseeable, but lamentable, problem is the fact that some people change their minds after agreeing to participate. Whether it’s just too much trouble, they start thinking it’s like Big Brother watching their every move or it’s a poor fashion accessory (see below), some people opt out of participating in the PPM after a short time. This led to some big problems over the summer.
On the other hand, some people are very diligent about participating. The limit for participants is 2 years (!) on the PPM panel. Unfortunately, a problem has surfaced in Houston where some people have hit their 2-year limit, yet Arbitron has not had enough replacements for these forced retirements.
Participation rates still suck – A new way of data collection doesn’t mean more participation. Every type of market research is facing a crisis with respect to getting people to agree to participate. Even though Arbitron is including cell phone only households, they’re finding it difficult to maintain their sample sizes, especially in the dreaded 18-34 demographic.
This is especially alarming when you consider that the size of the PPM panel over the course of a “book” is much smaller than the in-tab in the diary methodology. There are fewer individuals in the sample for a 12-week period, but since they are (theoretically) participating over a longer period of time, the number of persons-days tracked is larger. However, if the participation rates plummet, you have the same weighting problems that we have with the diary method.
Pagers are no longer hip – Beyond the problems with getting people to agree to participate, Arbitron is also presenting respondents with a fashion dilemma. Who wants to look like a dork by carrying a pager? In fact, Arbitron has admitted as much. In one of their conference calls they pointed out that one of the reasons for “voluntary noncompliance” with young women is that they didn’t want to carry it while going out.
In other countries, electronic measurement has meant different technologies. In some countries, such as Switzerland, they’ve used watches, but the consensus seems to be that a cell phone, the most ubiquitous electronic device around the world, will ultimately be the best bet.
Men are back – Male listening levels have gone up noticeably so far, which accounts for the stronger showing so far for rock-based formats. In fact, the early results show that men account for more quarter hours than women and that they account for more tune in occasions.
At work networking – The results so far are unequivocal, employed respondents listen to much more radio than those who aren’t. The latest figures I’ve seen are an average of 16 hours a week for workers vs. 9 hours for those who don’t. At-work listening has always driven the success of certain formats, such as classic rock and AC, but all formats have to be aware of the disparity in listening. This dependence on at work listening is leading to a dramatic quarter hour drop-off on the weekends. So far, the data shows that weekend quarter hours are more than 20% lower than during the weekdays. The interesting information will come from analyzing minute-by-minute data to find out how much button pushing goes on at work as compared with in-car or at-home listening.
Programming tactics still matter – The evidence so far shows that certain programming, promotions or giveaways can spike listening, just as we’ve always suspected. But we’re much more likely to find that things need to be simplified. Let people know about the benefit – move the little details onto your website – and they can make up their minds about whether they want to listen. We have to find compelling content, then we need to make sure we let listeners know what they get out of the deal.
These changes are exciting, troubling and inevitable, all rolled into one. New York just went into pre-currency, with LA, Chicago and Riverside to follow early next year. These markets will give us a wider geographic and demographic spectrum, so in the next few months we should see if these trends will continue to be the hot topics for PPM.
Pat Welsh, Senior Vice President/Digital Content, Pollack Media Group, can be reached at 310 459-8556, fax: 310-459-8556, or at Hq@pollackmedia.com.
