Earlier this month, Fox became the first network-owned station group to agree to license Rentrak's digital set-top audience ratings for all of its stations. The deal covers 28 Fox-owned stations in 18 local TV markets from New York City to Ocala-Gainesville, Fla.
If Nielsen fails to come to terms with Fox by the end of today, it would be the first time since Arbitron operated a competing local TV ratings service that a network-owned stations group agreed to do business exclusively with a ratings provider other than Nielsen. Arbitron left the local TV ratings business in the fall of 1993, and was acquired by Nielsen last year and now operates under its Nielsen Audio banner.
While Fox would be the highest-profile station group to walk away from Nielsen, others have already done so in selected markets. Two years ago, the Sinclair Broadcast Group went exclusively with Rentrak for ratings in four of its local TV markets.
While other major station groups have also signed deals to license Rentrak data -- including a recent multi-market agreement by the CBS station group -- most of those deals have been more of a hedge to utilize Rentrak's data to augment conventional ratings contracts they have with Nielsen.
Unlike Nielsen, which utilizes either paper diaries or combinations of paper diaries and electronic meters in all but the biggest TV markets, Rentrak utilizes a method that extrapolates TV audience estimates directly from digital TV set-top devices consumers use to watch television. While Rentrak's methods are not without question -- the Media Rating Council recently required Rentrak to stop using the term “census-based” to describe its ratings -- local TV stations are less than happy with the quality of some of Nielsen's local TV measurement methods, which are considered creaky at best.
Nielsen also created some tension with local TV stations when it announced plans to begin introducing “broadband-only” homes -- households that receive no subscription TV services of any kind -- into its local TV ratings samples, though it backed down from that plan after the National Association of Broadcasters complained about the move, although Nielsen is adding broadband-only households to its national TV ratings samples.
While much of Fox’s conflict with Nielsen relates to its ratings methods, executives familiar with the negotiations say there is also a considerable bottom-line aspect to it. They characterized the price increase Nielsen is seeking to renew its local TV ratings contracts as being exorbitant. They said the Fox stations also feel they can switch to Rentrak’s ratings with relative impunity, because Rentrak is believed to have highly regarded audience data for two of the most important advertising categories sold by the local Fox stations: automotive and entertainment.
When CBS announced its non-exclusive deal with Rentrak, it explicitly cited the opportunity to utilize its data on automotive consumers as part of its rationale.
Nielsen and Fox have gone toe-to-toe in the past. A contractual and methodological tiff between the two companies a decade ago led to the first Congressional hearings since the TV ratings scandals of the 1960s, which was the basis for forming the MRC as an industry self-regulatory body. While terms of Fox's eventual contract renewal with Nielsen were never disclosed, Nielsen made a number of significant methodological changes following that campaign, including weighting its ratings samples to give greater credit to African-American and Latino viewers.
“Nielsen is committed to providing comprehensive and representative television measurement for our clients and the industry,” Nielsen said in a statement provided to
MediaDailyNews this morning. “We have a long history of working with our clients to offer best in class measurement and currency based products upon which the media and advertising
industries can transact with confidence. Though we do not comment on contract negotiations publicly, Nielsen continues to work closely with Fox and all clients to meet and exceed business
Wayne Friedman contributed to this story.
Obviously the two parties may have very good reasons for being at loggerheads on the rating service subscription, so it will be really interesting to see what happens if the Fox stations drop Nielsen for local market audience data. This could well be a test case for the "Big Data" proponents, as, for the first time, buyers and sellers will see what differences are actually demonstrated in household tune in levels on a show by show basis. At the same time, buyers will be exposed only to household ratings, not viewer ratings, which means that planners may have to rethink the way they state their GRP goals and develop new sets of norms to guide them. Also problematic will be attempts to target demographically as household ratings do not tell you whether your consumer target actually viewed a telecast when the set was tuned to it. Overall, a typical adult is present before the set only about half of the time when it is in use. Moreover, this factor is not a constant. In younger, larger households the younger adult may be watching only 35-40% of the time because there are more residents----and visitors----who may actually be using the TV set. In smaller households---with one or two members, set usage correlates far better with adult viewing. A similar dichotomy exists between affluent and less affluent homes as the former also tend to have more residents while the latter are often headed by older folks.
Ed, my recent in market experience proves that what you are suggesting is true even for those who still use Nielsen. On a majority of brands, using Nielsen age gender numbers resulted in buys that were targeted to brand purchasers/ likely to buy/ pick any segment you want worse than if programs had been selected statistically randomly. On the others, it was only slightly better than random selection. That is using national Nielsen. Using local is even uglier. Using STB and all kinds of HH information we drove brand sales by double digits. Just because Nielsen tries to count viewership and reports it doesn't mean it is even close to an approximation of reality.
Jon, I think that often, "we", myself included when I was in the agency business, have asked Nielsen to give us more data than it is capable of supplying. A case in point, is commercial ratings. Do we really think that the system now in place----or, for that matter, any electronic system-----can really tell us whether people are "watching" commercials?
Your exchange reminds me of the 4th Chapter in Lewis Carroll's "Through the Looking Glass."
‘I know what you’re thinking about,’ said Tweedledum: ‘but it isn’t so, nohow.’
‘Contrariwise,’ continued Tweedledee, ‘if it was so, it might be; and if it were so, it would be; but as it isn’t, it ain’t. That’s logic.’
In due course, if Nielsen has it's way and wiser heads do not prevail half of Nielsen's reported national data (NTI) may be modeled - not tabulated. Then, what will we all have?
Instead of arguing about historical labels and false precision, the industry ought to consider what happens, if and when, there is no significant difference between Nielsen and Rentrak.
It should also be noted that if advertisers and agencies had been full partners in defraying the cost of media research, perhaps the industry would not face the Faustian bargain it confronts today.
Other modern countries have better, more equitable, ways of allocating the cost of media research.
Onwards and upwards.
Nick, your Lewis Carroll remark aside for now, when you say that half of Nielsen's reported national data may be modeled, not tabulated, could you elaborate? What kinds of data are you referring to-----out-of-home, new platforms, product usage? Also, if Rentrak and Nielsen do, indeed, produce virtually the same household ratings on a show by show basis----which I'd be surprised to see-----this raises the question about how fine one can slice and dice the info, which gives Rentrak an advantage, one would think. Finally, you imply that we would all be better off if advertisers and/or their agencies had paid their fair share of the costs of Nielsen's audience research from the beginning----in other words half of the cost. That's all well and good, but who is to decide what's a fair price for Nielsen ratings ---a consortium of agencies, advertisers, the TV stations and networks as well as Nielsen? Also, who gets to decide whether Nielsen turns a profit and how big a profit? Who gets in on the forward planning process as Nielsen, quite legitimately, plots its future course----like buying Arbitron or launching new online initiatives? Should advertisers have veto power over such decisions because they pay half of the costs? I'm not so sure that what "works" in other countries is an ideal model for the U.S.
As I carry a brief only for research quality,
my advice would be for you to ask Nielsen directly about a "secret" project, code-named "NPX."
From what I read in the press, here's the story to date:
Nielsen's avowed goal for NPX is to reduce the national sample error by half -- at the lowest cost possible.
That is laudable on its face.
Nielsen is wearing a mask that conceals both its true purpose and the probable outcome.
Nielsen does NOT plan to reduce sample error by expanding the size its existing national people meter (NPM) panel -- a methodological "best practice" in survey research for as long as there has been a statistical science.
Nielsen intends to diminish sampling error by modeling demographic data from set meter-only households and folding them into the their tabulated national panel data in very large numbers.
The resultant data may -- on a technicality --have a somewhar smaller sample error, but we would only be fooling ourselves if we accepted such ersatz ratings data as better television audience measurment.
In reality, it is statistical horseradish. Easy to prepare, but too hot to use.
It's time for transparency, wisdom and hard work to prevail. No more quick fixes. This may be our last chance to get the best media measurement in the world right.
We'll see who cares.
ONWARDS AND UPWARDS.
Upon review, it appears that Nielsen only intends to double the size of its national panel,
NOT reduce the sample error by half.
(That, of course, would require that Nielsen quadruple the size of its national panel.)
I regret the confusion my error may cause.
As before, best to consult Nielsen on their plans for the future -- as well as the MRC (Media Rating Council) which is responsible for the Congressionally-mandated accreditation process to which all currency measurements should be subject.
Thanks for your reply, Nick.
The plan you outlined resembles some discussions I had with interested parties a few years ago where the issue of "marrying" data from several sources came up, dealing more or less with the same subject. The problem at the time was how does a TV rating company whose business plan allows it to maintain a panel of only 20,000- 25,000 homes, integrate its TV audience findings with a far larger database containing other ad-relevant information, namely product usage, brand sales, etc. data? The goal:to break out TV shows, show types, networks and cable channels, etc. in terms of their ability to reach heavy product users, brand buyers, etc. Since a typical TV rating, across all dayparts and network types is now on the order of .3-.5%, it is virtually impossible to get reliable product/brand usage breakdowns for a majority of shows when their average sample size per telecast, using a 25,000 home panel, is only 90-100 "respondents. Breaking this down by heavy user groups that amount to only 10-20% of the total population for even the larger product categories, or for brands with a share-of-purchase rate of only 3-10% is virtually impossible. Like it or not, the only solution we could come up with-----and not the ideal one, by any means----was for the findings of the much larger marketing panel to be married with the TV rating panel's audience data using what used to be called "simulation" techniques. Since there is little hope that advertisers will step in and foot a large part of the bill for a huge expansion in the TV rating sample, with these homes drawn from the marketing panel, simulation might be a solution, providing it is done very carefully and, equally important, providing it is validated by an independent panel of research/marketing "experts". While I share your concerns about such a process, I hope that whatever Nielsen is contemplating passes the "is it reasonably accurate" test.
Thank you for sharing your experiences, Ed. It is curious to see how little progress the parties have made over time despite the good counsel from people like you.
Based on Logic 101, your final test for Nielsen is best classified as "Necessary But Not Sufficient."
Saint Bernard of Clairvaux wrote around 1150 AD: "L'enfer est plein de bonnes volontés et désirs."
(Hell is full of good wishes and desires.)
Recent global history tells us Saint Bernard is still right.
The U.S. National Bureau of Economic Research tells us the Second Great Depression began in December 2007.
The majority report of the U.S. Financial Crisis Inquiry Commission reported its findings in January 2011.
CONCLUSION: "the crisis was avoidable and was caused by:
Widespread failures in financial regulation, including the Federal Reserve’s failure to stem the tide of toxic mortgages;
Dramatic breakdowns in corporate governance including too many financial firms acting recklessly and taking on too much risk;
An explosive mix of excessive borrowing and risk by households and Wall Street that put the financial system on a collision course with crisis;
Key policy makers ill prepared for the crisis, lacking a full understanding of the financial system they oversaw;
and systemic breaches in accountability and ethics at all levels.“
You said it earlier in the day "we ... have asked Nielsen to give us more data than it is capable of supplying."
It is more than time we came to terms with reality. The penalty period is about to begin and how we've played to date will cease to matter.
Those dependent on valid, reliable and useful data need a winning "Gooooooooooooooooool" before regulation time runs out.
Onwards and upwards.
To my mind based on what I've read, there are 13 million dishnetwork subscribers. of those approx. 960,000 watch fox news. Why in the hell should the 12 million who don't watch their propaganda pay a higher monthly fee just so fox can be back on? Leave them off Dish, and get them off Direct TV too!
Where is the campaign for ala carte programming? Pay for the channels we watch. We should be able to buy what we want, not have garbage forced onto us.
Given the fact Fox brags about ratings and having the most watched with 2.4 million watching on average, that 960,000 represents a HUGE number of their viewers...Just keep writing Dish and tell them not to give in, keep fox off, numbers matter and those who don't want to watch it, shouldn't have to pay for it!!!
Ben, you're totally off-topic here - this is about Fox TV Stations (network O&O affiliates - think Simpsons, Sleepy Hollow, NFL on Fox)not Fox News. May as well correct a few misconceptions though.
re; Fox News/DISH - I don't know where you got the 960k #, but it sounds like a cume, whereas the 2.4 million (if accurate)sounds like the average. Subtracting a cume from an average is not the correct math to estimate the impact of the loss of DISH on Fox News ratings. Going by your #s it would likely be around 200k potentially lost (avg). As far as letter-writing goes, I don't know the #s but the ones DISH is getting in favor of bringing Fox News back are undoubtedly orders of magnitude higher than any suggesting they keep it off.
With respect to a la carte in general, it's very much a hot topic in the context of OTT but I'd like to suggest it's not a simple question. What people pay in a bundle for stuff they don't watch subsidizes the whole ecosystem, unbundling will lead to people a/getting less content, and b/IMO when you think through the scenario, paying a total bill for "only what I want" that will often be HIGHER than the bundle. ESPN is a good way to illustrate this..the average MPVD ("cable") customer is said to pay around $6/month for ESPN but research suggests probably only about 25% of those value it enough to continue to pay a la carte if given that option. That's a very large niche but if everything went a la carte and 75% of their "subs" left, they'd have to quadruple fees to the remaining 25% or take a big haircut on their topline (and also reduced ad $ due to smaller reach). ESPN then would need to charge nearly $30 to break even - and how many would drop out at that point because they couldn't afford it? If ESPN could only get away with charging say $15 their revenue would plummet and this would soon affect a much larger ecosystem..viewers, advertisers, leagues (reduced rights fees), in turn player salaries, etc, etc.
The average viewer watches about 15 channels regularly, and a bunch more periodically. Imagine hitting a toll (subscribe or pay for on-demand) every time you say to yourself "hey, I want to see that".
I am not saying the traditional model is not under a microscope and likely to see significant evolution over the coming years. Technology is forcing the hand of everyone involved. But technological capabilities do not suspend laws of economics. There really is no free lunch at the end of the day. You can get 1/a smorgasbord of premium content for a flat/bundled rate, or 2/you can have precisely the content you want, only what you want, when you want it, how you want it (device), anywhere you want it, ad-free. I just don't see in the end how you are going to get both.
This is all merely my personal opinion but I haven't seen many game this out completely yet - the numbers don't add up when you do.