Tom Birch looks at the new ratings battle

0

When it comes to the radio ratings business, Tom Birch knows what he’s talking about. He founded Birch Radio, sold it to VNU (now known as The Nielsen Company) and continued to run the radio ratings company until 1990, two years before VNU shut it down. Naturally, we thought he’d have a lot to say about Nielsen getting back into the radio ratings business to challenge Arbitron – and we were right.


RBR/TVBR Executive Editor Jack Messmer (JM) asked Birch (TB) what he thinks about the move by Nielsen.

TB:    I think that’s it a very healthy thing for small market and mid-size market broadcasters. But Nielsen doesn’t go into a market just to do 50 smaller markets.  If they’re going to address this market, you’ve got to be ready for them to move up the market.  I’ve been a proponent of telephone methodology for years – and telephone carries the advantages of  lower cost, higher response rates – but keeping a diary methodology is a smart business move because now Nielsen’s product will look very much like Arbitron’s, which makes it very easy for agencies to go back and forth between the two services.   That kind of portability will enable Nielsen to move into markets if Arbitron sticks to guns and goes into markets like Raleigh, for example, with PPM.  Broadcasters that don’t want to shell out a million-plus dollars a year for audience ratings will turn to Nielsen.  So I think the headline in this is that this entrance by Nielsen with the diary methodology into these 50 smaller markets portends, I think, a limitation on how far Arbitron will be able to extend the PPM.  I would not be happy if I was at Arbitron right now.  They’ve got a tough couple of days.

JM:    What do you think of the sticker diary idea?  Do you have any idea as to how that works?

TB:    I don’t really. I’m not familiar with it.  All I know about it is what I’ve read in the trades and just from what I’ve read it sounds like a very novel way to make it easier for people to keep the diary.  The less intrusive you can make a methodology the higher the response rate will be.  Apparently Nielsen has had success with this internationally, but I really don’t know much about it.  I think it’s still, even for Nielsen, a fairly new innovation.

JM:    I guess they’re really rolling it out on a large scale just this year in Australia.

TB:    It makes sense.  What I find much more intriguing is the address-based sampling technique.  I think that makes a whole lot of sense.  That sampling methodology, you know in choosing members of the sample, that could possibly result in differences between Nielsen’s diary ratings and Arbitron’s diary ratings.  I think that, more than anything else, could drive some differences.  But my prediction is that the ratings in these markets, between the two services, are going to look very, very much alike.  The only criticism I would have of it is the lack of frequency.  Only doing one measurement a year – and what really startled me is doing it over just and eight week period of time – I think is going to put Nielsen at a significant disadvantage in competing against Arbitron.  I haven’t looked closely, but I think some of these markets are actually continuously measured by Arbitron, and if not continuously measured they’re I think virtually all of them have Spring and Fall surveys.  So if a broadcaster were to sign up for the Nielsen survey in the spring, that survey almost by definition will become irrelevant as soon as the Arbitron data is issued in the fall.  So that broadcaster is going to find himself being beaten up by agencies that are using the Arbitron data and he’ll be put into a position to having to reconsider subscribing to Arbitron.  I think it’s going to be an interesting test to the wills of the broadcaster that sign up for just one survey a year.

JM:    How is your survey company going?  You did your one market. Is that being used a lot by your agencies in the area?

TB:     We’re getting tremendous value out of it internally.  I have not made an effort frankly to market what we’re doing beyond our area.  I’ve made a couple of phone calls and there are a couple of people I think are interested in possibly working with us next spring, but nothing has been finalized.  I’m still sticking to my day job of running a radio group.

JM:    Are you doing that one once a year or twice a year?

TB:    I have the luxury of being able to do it once a year, and we’re going to do it again in the spring of ’09, but we’re a very small market and I think in very small markets once a year is fine.  But my earlier comment on the Nielsen study, when you’re in a market where Arbitron is doing two studies a year and you come in and do just one, I think you’re going to have a problem with it.  If I could afford it I would do it twice a year, but where we are there’s very little competitive activity.  We don’t have format changes and things happening that would require measurement more than just once a year.  Frankly, even one a year may be overkill in very small markets.

JM:    Of course Nielsen is pitching the size of the sample as the counterbalance to the number of times a year because their sample is like 60 to 100% bigger than what Arbitron is doing in the same markets.  How do you balance that?  How important is sample size versus frequency I guess?

TB:    Sample size is very important.  I guess the way I would counter that is that I don’t think that it’s necessary for the sample to be so large that it supports demographic cells within day parts.  For example, I don’t think that people should be looking at 18 to 24 year-old male listening in afternoon drive.  I think that the better compromise is to keep the sample size within an affordable range, but limit the amount of data your produce from that sample.  For example, we don’t, in our study, we do not allow our salespeople to talk about anything other than the aggregate demographic combinations like 25 to 54 and we don’t focus on individual day parts, but rather on day part combos like 6A to 7P and 3P to Midnight, because I think anytime you go beyond that, even if you have a sample size that’s double what Arbitron currently has in these markets, I think you’re still going to have a great deal of error around the estimates.  The word that people keep forgetting when they use radio ratings is “estimates,” that’s all these are.  So the broader the groups that you can look at, the better, and I know that flies in the face of what PPM is trying to do in the major markets, but the reality is that unless you have some kind of an electronic measurement device that is reporting from a sample with a response rate that is in excess of at least 20%, I would hope 25%, you’re not getting reliable estimates.  I may sound like a Luddite, but I believe you should use large samples, as large as you can afford, and be conservative as to how much data you try to ferret out of the survey.

JM:    You can do that with your own salespeople, but you can’t control what the agencies then do with their software that interprets the data.

TB:    Unfortunately, I’d say that is true. One of the things that I found interesting when Cumulus put out their RFP last spring was a plan that the software that would be delivering the data contains governors in it to prevent abuse of the data.  I don’t know whether or not Nielsen is going to be able to address that.  Certainly as soon as they issue their tapes and they send them out to the third party processors there’s very little they can do to prevent people from trying to read more into the sample than the sample can support.

JM:    I think 2010 will be very interesting. Once they get this up and running in 2009 I’m sure that’s not going to be the end of it so.

TB:    Oh, I promise you that. I think the first thing to look for is an expansion of beyond just one survey per year.  I think that will happen rather quickly, and I think the second thing to look for is for them to be moving up, not down, in terms of market size.

JM:    I guess the big hope would be that we don’t end up with a monopoly by a different company.

TB:    I wouldn’t look for that to happen anytime soon.  I think that Arbitron is very, very entrenched.  They are a world class media measurement company.  They’ve proven over the years – I certainly know it first hand – to be very, very competent competitors.  What really is interesting is the open warfare that I think this unleashes now.  The gloves are off and I think now Arbitron, with PPM, may be eyeing the television market, so you could see us going back to where we were in the ’70s when Arbitron was measuring both radio and TV.  Of course it’s been since the early ’60s that Nielsen was in radio, but it should be very interesting, especially from your point of view writing about it.