Stations On Edge Over Nielsen’s New Ratings

Results from the ratings service’s new “code-reader” viewer measuring system that is set to replace the age-old paper diaries become available in June and what the data will look like is anybody’s guess. Uncertainty over the new system, its accuracy, its impact and its pricing appears widespread in the 14 markets where the new system is being introduced.

Will their ratings go up, down, or stay the same?

In a nutshell, that’s what broadcasters in 14 markets are anxious to find out as they await the arrival of the first sets of new data to be produced by the Nielsen “code-reader” viewer measuring system that is set to replace the age-old paper diaries.

The new data is almost here, due to be delivered in June to stations in the three markets where the system was first operational — Flint-Saginaw-Bay City, Mich. (DMA 70), Madison, Wis. (82) and Reno, Nev. (107) — and in July to stations in the remaining 11 markets where the system is also operational and undergoing testing.

What the data will look like is anybody’s guess. John Humphries, GM of Hearst Television’s WGAL Harrisburg, Pa., an NBC affiliate, admits he’s apprehensive, but also optimistic. Harrisburg (DMA 45) is one of the 11 markets due to receive the so-called “impact data,” designed to be compared with ongoing diary-based data, in July.

“Nielsen’s been measuring this market for 50 years [and] we also have done proprietary research over the years that use different samples and different methodologies,” Humphries said. “And Rentrak is in this market too. While we don’t subscribe [to Rentrak], we have seen some of those numbers.

“And every one of those independent research sources tells the same story about my station, so I would be shocked and, as you might imagine, disappointed if the new code reader methodology told a different story,” he said. “So, apprehensive, yes, but also optimistic that we’re going to retain our place in the marketplace.”


In addition to Harrisburg, the other second-wave markets are Grand Rapids-Kalamazoo-Battle Creek, Mich. (DMA 40); Fresno-Visalia, Calif. (54); Albany-Schenectady-Troy, N.Y. (58); Mobile, Ala.-Pensacola (Fort Walton), Fla. (59); Tucson (Sierra Vista), Ariz. (71); Paducah, Ky.-Cape Girardeau, Mo. (81); Charleston, S.C. (95); Traverse City-Cadillac, Mich. (118); Yakima-Pasco-Richland-Kennewick, Wash. (DMA 122); and Santa Barbara-Santa Maria-San Luis Obispo, Calif. (123).

Uncertainty over the new system, its accuracy and its impact appears widespread in the markets where the new system is being introduced. “We have been slow to embrace this next step for Nielsen,” said Chuck Roth, director of business administration for Quincy Newspapers, owner of ABC affiliate WKOW Madison.

The station only recently agreed to participate in the code-reader trial, Roth said. “There have been a lot of stops and starts and we have had concerns about how they were moving forward.”

However, many of Quincy’s concerns were addressed at a meeting with Nielsen executives at the NAB Show in Las Vegas earlier this month. “We left that meeting cautiously optimistic,” Roth said. “We felt that they might finally be at the point where their code reader system might work. We felt fairly satisfied that they had a good plan on how they were going to deal with the demographics.”

But it’s the demographic aspects of the new system that have raised the most questions among broadcasters.

The new Nielsen system consists of two parts. One is a new generation of passive code readers that track viewing in Nielsen households. The boxes, which are affixed to every TV set in a household, identify shows by “listening” to the audio and then identify the broadcasting station via a Nielsen “audio watermark” embedded in the signals.

Unlike Nielsen’s people meters, the passive code readers do not know which member or members of the family are doing the viewing, so demographic data is to be collected from both local and national people meters located in the closest larger markets and combined with the passive data from the code readers.

Nielsen believes this system of collecting demographic data – the other, second part of the system that Nielsen refers to as “viewer-assignment” data – will produce household demographic profiles that will closely resemble the demographics of the code-reader homes. Nielsen also intends to use the system to replace diaries for collecting demos in its 31 set-meter markets (DMAs 26-56). In those markets today, the hard-wired set meters generate the household ratings, while the diaries produce the demos.

Broadcasters’ reactions to viewer-assignment demos range from cautious to downright disbelieving.

Some broadcasters say they’re reserving judgment until they see the test data. Without the data, “it’s really hard to tell” if the viewer-assignment method is viable or not, said Billy McDowell, VP of research for Raycom Media.

But at least one owner of small-market TV stations — Bonten Media — recently dropped Nielsen in favor of Rentrak due to concerns over pricing, accuracy and the way Nielsen proposes to collect demographic data in smaller markets.

In an interview with TVNewsCheck posted Monday, Bonten CEO Randall Bongarten called the new demo-collection method “ridiculous.”

“They have come to this recognition that the demographic data in particular is extremely unstable and so what they’re going to do for these smaller markets is they are not even going to collect demographic data,” Bongarten said of Nielsen.

“So basically what they’re doing is assuming that whatever the position is of the station in that bigger market is the position of the station in the smaller market. You’re applying the same exact demography. That’s ridiculous. It really doesn’t apply because the demography can vary with the strength or weakness of the individual station in that market,” Bongarten said.

“We went to Rentrak and I am extremely happy with them,” he said. “It’s a much more stable number, much larger sample, and we get ratings every day. It’s really, really superior to Nielsen and it’s a lot less expensive.”

Broadcasters say they’re uncertain about the costs of Nielsen’s new measurement system and are bracing themselves to be charged more than they’re paying now for paper-diary measurement. One station contacted by TVNewsCheck paid $12,000 for the encoder used to participate in Nielsen’s code-reader test.

“[Nielsen has] been very quiet about [pricing],” said one local broadcaster, Tom Bier, general manager of Morgan Murphy’s CBS affiliate WISC Madison.

Nielsen wouldn’t comment on how it plans to price the new service. But Matt O’Grady, Nielsen’s EVP-GM of local media, expressed confidence that the data produced by the new measurement system would produce new revenue for stations. “We need to ensure that we deliver solutions to [stations] that they can afford and that accurately represent the market and bring more dollars into the market,” O’Grady said.

Whatever the costs, broadcasters seem to be hopeful that the new system will yield better data than the old one. “Generally speaking, we are looking at this as a good step for Nielsen. There has to be a better system than what they have today,” said Quincy’s Roth.

“I can’t [imagine] that this [would not] be much better, much more reliable [and] much more consistent,” WISC’s Bier said. “I think it’s going to serve the industry well when we get to that point.… Here in Madison, we are happy to be among the first adopters of this system.”

Comments (16)

Leave a Reply

mary lawrence says:

April 29, 2015 at 9:44 am

Any ‘new’ technology will be paid for by the stations.

Kristine Melser says:

April 29, 2015 at 10:32 am

So, their answer to improving the small market diary system from the 70’s is to collect set top HH data and use neighboring market demo data … because the demographics of small markets and large/medium markets are so similar with similar lifestyle values. Has Nielsen ever been to Rapid City, SD and Minot, ND ? One visit there and you will know the age demo’s are different … let alone lifestyle and values.

Paul Beecher says:

April 29, 2015 at 11:08 am

Rentrak (and any other nascent entrant into the set-top box tracking methodology) is the clear disrupter to the ancient Nielsen Empire. The industry can only hope that Rentrak never “sells out” to them (as Arbitron did.) We need competition. Competition drives innovation which in turn drives better service and better prices. The paltry sum of paper diaries that Nielsen attempts to secure relative to the sheer number of Rentrak STB’s providing real-time viewing is incomparable. Yes, there’s a general number/percentage of HH’s that must be secured to provide a “statistically reliable
sample (give or take a few points.) But how can you argue with real-time data? Furthermore, utilizing adjacent larger market demos and attempting to relate those demos to smaller outlying markets is beyond credible. So, Albany & Binghamton people are just like New York City people? Peoria and Quad Cities people are like Chicagoans? Bakersfield and Yuma people are just like LA people? Please.

    John Bagwell says:

    April 29, 2015 at 11:47 am

    I don’t disagree with your criticism of Nielsen, but Rentrak has many flaws as well. People just look at the sample size and think it is perfect.

    Roxanne Tedeschi says:

    April 29, 2015 at 1:53 pm

    Nielsen is operating a megbucks business by providing protection for it’s Network clients and their local affilates. IF a station does not subscribe to Nielsen, Nielsen is not obligated to provide accurate data for that station, and usually assign them a very small number of viewers of hashmarks….
    I know, I have lived it many years and then turned on Rentrak. There is a world of difference.
    Also, Demo data can be gotten if Nielsen wants to pay the price for the viewer’s TV box data.
    Imagine if Nielsen placed its new system in New York or LA or Chicago…..
    What a joke

    Wagner Pereira says:

    April 30, 2015 at 1:23 am

    Anyone who thinks a Company with a valuation in the Billions will risk their credibility by making up ratings for a station who are not subscribing is a fool who needs to start wearing a tin foil hat.

    Wagner Pereira says:

    April 30, 2015 at 1:39 am

    @GuyInTheKnow – 9 different STBs are active on 16 different channels in my house right now – and no one is watching any of them. Great real time ratings using that data. If you want to see how bad STB data is, look at the “Hot” lists from FiOS, TiVO, DirecTV or any of the others. As I type this at 1AM, according to FiOS real time data from STB, the #1 Show is Dr. Oz (rerun) and #2 Wheeler Dealers on Velocity (ever watch that channel or show?) #3) Showtime 4) Some unknown PBS show 5) TruTV 6) Cinemoi airing the 1969 Movie Cactus Flower 7)ActionMax 8) CSPAN3 9) Community Access Channel 10) Fox Sports 1. Anytime Rentrak would bring me that ranking i would laugh them out of the office. Even more interesting not 1 of the Top 20 FiOS claims is in the Top 20 from TiVO (which has Bloomberg at #1 – even though there is an infomercial on it at this time). Great Real Time Data.

Susan Durant says:

April 29, 2015 at 11:20 am

so basically they are going to make up the demo…. wow… no wonder digital ad sales are hurting local stations!!!!!!!!

    John Bagwell says:

    April 29, 2015 at 11:49 am

    Rentrak basically makes the up the demo ratings as well.

    Kristine Melser says:

    April 29, 2015 at 3:51 pm

    Digital makes up the impressions as well thanks to compu bots. Digital is just a buzz word right now profiting from buzz, popularity and the luck that no advertising is really transparent right now delivering what you pay for.

    Nate Mann says:

    April 29, 2015 at 6:45 pm

    …digital is just a buzz word… right… now? Woah. Stupidest comment ever posted on the Internet ever. And that’s saying something.

Doug Halonen says:

May 1, 2015 at 7:34 pm

I would love to see the data that convinced Nielsen that a nearby larger market can act as a reliable demographic proxy for the nearby smaller market.

Paul Beecher says:

May 4, 2015 at 9:48 am

To Insider: In reference to your “hot list” of channels “trending” at 1am. I actually believe those channels you listed are, in fact, accurate. As a subscriber to Verizon FiOs (and an active curiosity seeker of viewing trends from this trending feature) I find the reliability to be dead-on. The software pulls real-time data on pure STB tune-in. Is it possible the STB could be “tuned-in” and no one is actually watching? Of course. But the same goes for any research methodology or technology. Hell, if I ever had to fill out a diary (and I haven’t) but had to “recall” what I watched on Sunday night at 7pm, I’d most likely say “60 Minutes”. But was I fully engaged for the entire hour? No. Some segments you watch, some you don’t. I can easily flip to alternative niche channels of interest (including Wheeler Dealers, CSPAN, Bloomburg or PBS, if I choose.) That’s the beauty of choice – and the challenge for broadcasters and programmers – to get better at “engaging” and holding their potential audiences.
BTW Insider…if you really have 9 STB’s at your residence (and wow, the cable or satellite guys love you) how do those 9 boxes tune in to 16 different channels?

    Wagner Pereira says:

    May 5, 2015 at 2:40 pm

    @GuyInTheKnow. Let’s answer your last question first. 9 STBs tuned in 16 different channels because DVRs have 2 to 6 tuners in my system. I only counted the channels 1) showing or 2) recording. As for the list being accurate, I have seen evidence it is not. One only needs to compare the DirecTV hot list with the FiOS hotlist and one can see the difference. But regardless, the FiOS data, even if no one was watching the STB in the middle of the night as you suggest (even though they now default to power to standby), then again, it proves my point of Rentrak STB data as not being accurate. Nielsen wired houses have safeguards against that (as well as who is watching).

More News