In Part I of TVNewsCheck‘s three-part special report on audience measurement, Media Ratings Council Chairman and Raycom Media chief researcher Billy McDowell discusses the state of local TV ratings and what’s being done to make them better. He also gives no ground in the battle over what measure should be used in local TV: live-plus-same-day is the only way to go.
TV ratings are not like the weather. If you complain about them, somebody will do something about it — or at least try to. That somebody is the Media Ratings Council, a 48-year-old group of media companies (electronic and print) and media agencies that monitors, audits, accredits and otherwise tries to keep Nielsen and other measurement services honest.
The CEO is George Ivie who manages three fulltime and one part-time professionals in New York. And the current chairman is TV broadcasting’s own Billy McDowell, VP of research for Raycom Media, operator of 36 full-power and two low-power TV stations in 36 markets.
Last week, while in New York for the annual meeting of the 132-member MRC board, McDowell took time to discuss with TVNewsCheck Editor Harry A. Jessell the state of local TV ratings, including Nielsen’s efforts to better them, the value of ratings based on set-top data and his no-compromise position in the feud between broadcasters and media agencies over the truest measure of local TV audiences.
An edited transcript:
Nielsen’s last big local measurement initiative was the substitution of local people meters for diaries in the top 25 markets. Is everybody now happy with the LPMs?
I wouldn’t characterize it as happy. There’s always issues. It’s been widely reported that there have been some sample size issues, that they’re limited. They’ve been able to increase the samples in New York and L.A. and that helps, but you’re always pushing the envelope when it comes to needing more samples in markets. That becomes expensive and hard to accomplish so we look for alternatives in that, too.
Other than the sample size, are there any other concerns with the LPMs?
None that I could really go into here, but we’re always monitoring what they’re doing. Individual markets have issues from time to time and they call on the MRC to help them out in interpreting it, understanding the methodology, suggesting improvements.
What are you hearing from Nielsen on what it’s doing to improve diaries in the majority of markets that still rely on them to one extent or another?
They’re still trying to figure out the diary markets. They’re working on a lot of improvements, some that we have suggested, some they’re doing entirely on their own. I know that sample size is an issue there as well. They have been public about their set-top box initiatives and we will see where it all goes.
I think that there will be diary measurement for some time and I think that we really do need to improve that, but not only the diary instrument itself. It’s really good at going into a local market and determining which stations are leaders in the market and things like that. Trying to predict what ratings are going to be at a quarter hour on a certain program in a certain demo is where it really becomes problematic. It’s too granular for the diary system as a whole. So if we could learn to use the diary information a little more responsibly.
In an earlier interview with Nielsen’s Steve Hasker, he said they are going to use set-top box data to supplement the diary information and come up with an amalgamation that’s going to be better. That’s essentially the idea?
That’s right [laughs]. I hope they buy it.
Why do you chuckle? Is that a disappointment? Would you have hoped that they would do something different? Do you have any better ideas?
Actually, we do struggle with that because not only are you looking for the perfect methodology to be able to measure all the nuances of what’s going on in a market, but you have to do that at a reasonable economic level so that stations can participate. We can come up with all kinds of grand ideas, but unless they’re affordable and you’re able to implement them, they’re no good.
In the hybrid markets — 26 to 56 — you get a mix of household meter data on a daily basis and the demo data from diaries during sweeps. I guess that is much better than just diaries, right?
Right. Actually the household meter is a pretty good tool because it’s one of those passive measurements. So the problem there is also sample size … and one thing that Nielsen is finding is that you want more stability in the measurement, but the other part of that is you need a proper representation of these markets … to make sure that they’re properly represented in the methodology.
Are there any special issues with those meter/diary markets?
Yeah. I can’t get into specifics, but again Nielsen is looking to improve all their measurements and there’s a lot of ideas being floated around right now. The set-top box information that you guys talked about is probably among the most imminent.
Let’s talk about set-top boxes. Rentrak is out there selling a local rating service based on set-top data. I did an interview with Marci Burdick of Schurz [last week] and she is pretty happy with it in Wichita and is going to try it in a couple of other markets. Is this valid viewing data?
Well, Raycom actually subscribes to Rentrak in two markets, Birmingham [Ala.] and Charlotte [N.C.]. I wanted to get a lot of experience with the data and so we have looked at Rentrak pretty closely. It’s really good in providing stability in measurement. If I have 60,000 homes in a market that are reporting data, it’s a lot more predictable. In the end, there’s not that statistical variation that you see in other measurements.
But there are a lot of people who are looking at them and wondering if they’re going to be able to properly represent an entire market. That’s where their real challenge is.
Well, for one thing, Rentrak isn’t measuring over-the-air-only homes, right?
Right and so there’s a lot of modeling that’s involved in Rentrak and whoever else is out there doing it. That’s where people really need to understand how it’s being done and get more of the comfort level with it.
Another problem I have heard is that set-top boxes stay on even when the TV sets are off.
That’s one of the other functions that they have had to model for. They actually have set-top boxes that can identify whether or not the TV is turned on or off and so they have to model to that. There’s a lot of modeling that’s going on with that whole system.
For the demos, too, right?
Well, they are modeling some demo information based on other sources of information they have. There are sources of information out there that people are trying to take advantage of and you take it and you model it and adjust it so that it makes sense. That’s a big challenge for a lot of us today.
In Charlotte and Birmingham, you still get the Nielsen numbers, right? How do the numbers compare?
The numbers are pretty consistent with Nielsen in most time periods especially in Charlotte. Birmingham is very consistent in most time periods. There are a few periods when they are quite different.
Are there any other set-top data providers trying to make their mark in the local TV market?
I know [Kantar Media Research] is. TRA Global is another company that’s out there. They have more of a national bent, but I believe they are working on local market initiatives now.
Have any of the come to you for accreditation?
Rentrak has completed a pre-audit phase and so we’re hoping they get audited soon. I am not sure what the status is of Kantar and TRA Global. I think they have had discussions with MRC, but neither has gone through the full audit yet.
Last year at about this time, MRC revoked its accreditation of Nielsen for the diaries. What was that all about and has that been remedied?
No. They’re still without accreditation currently, but that doesn’t mean that anybody has stopped using diary information. We will continue to work with Nielsen to seek out those improvements that make it worthy of accreditation again.
Is there anything in particular they need to do to get back in good graces?
I couldn’t be that specific about it.
Sure you could.
I could, but….
Do you see Rentrak or any of the other set-top services evolving into a real competitor to Nielsen?
I know Rentrak has more than a 100 local stations that are subscribing. So I am sure that that has got the attention of Nielsen, but we will have to see if they get out into the marketplace more. There’s a little more understanding about what their methodology is, how they’re putting together their model and whether or not they would really pose a threat to someone like Nielsen taking over.
Do you see Raycom following the lead of Schurz or Nexstar and dropping Nielsen, at least in a few markets, and substituting Rentrak?
Right now, it’s not a possibility because of contractual things, but we are constantly looking at the evolution of Rentrak and how it’s being accepted out in the marketplace. We have develop a good bit of understanding about their data and there are things we like, like stability, and there are things that we are a little nervous about, like the representation. Whether it it will become a total replacement for Nielsen in a market, I don’t know. But in Birmingham they have 60,000 set-top boxes. If they have that much information on my viewers, then I am interested and I want to see what those people are doing. It’s a good thing for me to have that additional data.
So you see it as supplementary for now?
Well, I know that a lot of markets have turned their Nielsen subscription off and I don’t know if that’s a possibility for Raycom or not. I couldn’t say.
You’re aware of the battle over DVR viewing, I’m sure. Nielsen came out with this live-plus-same-day and then the media agencies squawked that they didn’t want to use that data so they had to bring back the live-only and most will only buy on live-only. Is there a compromise here? Can we work this out like gentlemen and ladies?
I have been at some of those meetings and we have as broadcasters looked at the data closely. We believe the proxy for C3 ratings at the local level right now is live-plus-same-day. I mean it’s pretty evident that’s what it is and I know there’s been talk of getting just some kind of number that people can index to and say this is it, but that doesn’t account for the nuances of the markets or the stations of the time periods, of the demos or anything. Live-plus-same-day seems to be the best one at doing that. I understand the frustration of the agencies not being able to have C3 at the local level, but the samples, the methodology just won’t support it. So, from our standpoint, the best proxy is the live-plus-same-day rating.
OK, but the question is, can we meet in the middle somewhere?
If you’re talking about just using an artificial factor of some kind, that’s not a good solution because it doesn’t take into account all of the nuances in the way that live-plus-same-day ratings actually do.
We’re talking about a discount, right? You arbitrarily discount the live-plus-same-day ratings just to get a buy-in from the media agencies. Isn’t that better — isn’t that more — than just using live-only data?
Well, I think we will continue to demonstrate over time how effective the live-plus-same-day ratings are as a proxy and I think people will really understand after they have seen enough data.
Is Arbitron a factor at all in the TV side?
Not so much. I know they’re still going through some radio procedures there. So I think once they do that, I know that they talked to me about some out-of-home possibilities, but we have not done anything. I think we are waiting for them to settle their issues on the radio side.
Is there anything else on your plate that you would like to talk about here?
I’m just taking care of my stations and making sure that I have a good enough November book. It’s exciting how resilient television at the local level has been and how it continues to be the dominant medium.