Commentary
Nielsen Wants To Measure Netflix -- Does Netflix Even Care?
- by Wayne Friedman , Staff Writer, October 23, 2017
What does Netflix think about Nielsen's efforts to measure viewing of TV programming on its platform? Maybe just a shrug of the shoulders.
Netflix has long said that because it is not an ad-supported service, it doesn’t serve anyone to release viewership data — publicly, anyway.
Privately? That may be another matter. Major TV-movie studios/networks make deals with Netflix — and in return, sometimes, Netflix offers data of how those network programs are performing.
That’s only fair. Still, TV executives say that data isn’t much.
With the announcements of its SVOD measuring service — which will start by only analyzing Netflix — Nielsen provides viewership results of some Netflix shows.
For example, “The Defenders” pulled in 6.1 million viewers for its premiere episode during its first week of viewing. An episode of “Fuller House” hit 4.6 million. Those are nice, middle-to-low broadcast network numbers.
This is Netflix’s response to TV Watch concerning Nielsen's efforts: "The data that Nielsen is reporting is not accurate, not even close, and does not reflect the viewing of these shows on Netflix."
But what are we comparing? Full on-demand per episode access of TV programming (Netflix) in a given week versus live airings plus three or seven days of time-shifted viewing (TV networks).
Is that the same thing? And what value can you give, say to “engagement” of those Netflix viewers, who are spending around $10 a month, versus watching a similar show on an ad-supported network?
Apples and oranges. And in this new media world, there are different metrics and results analyzing various consumer interactions of media channels.
In the past, NBCUniversal and others tried to calculate Netflix programming viewership — all to remove the mystery and hype of what Netflix has become for many TV consumers and business executives. It hasn’t worked. Netflix continued to build its business.
For its part, Netflix knows exactly how many times its shows have been streamed — and when — from its own server.
By contrast, a broadcast network or cable network has never had this ability via traditional TV transmission.
And when Nielsen gets it wrong — especially with difficult-to-measure younger-skewing networks, like MTV or Nickelodeon, senior TV executives complain.
Netflix still doesn’t know specific demographics, exactly who is in the room during viewing. It probably has research to make good guesses. What about guessing Nielsen's efforts to measure Netflix? Maybe Netflix believes that's a house of cards.
Wayne Friedman is West Coast Editor of MediaPost. You can reach Wayne at wayne@mediapost.com.
Poor Netflix---sob, sob. Now it can't hide its viewing info anymore. As for Netflix claiming that Nielsen's data will be all wrong, that sounds like a lot of sour grapes---where's the proof, guys? Sure, Nielsen is just doing a survey, ,with all of the issues that go along with that----but why is Netflix so afraid of having the size and the nature of its audience measured? It doesn't matter a fig whether Netflix sells ads or not. It still attracts viewing audiences ---in competition with other content suppliers---and the industry as a whole has a right to know as much as it can learn about who's watching what----doesn't it?
Hard to blame Netflix for not releasing data since they are not serving ads and their only master is the consumer who votes with their monthly subscription. If and when selling ads becomes part of their business model, I'm sure Netflix will release all of the appropriate data advertisers require and expect.
And if they end up selling ads, "Goodbye, Netflix..."
The only people Netflix needs to disclose viewership to is its investors and its content partners.
“The industry as a whole” has no right to anything with Netflix.
Traditional TV is on its last breath and Nielsen is struggling to remain relevant even though it’s done as much as any company to hurt TV.
Welcome to the new world order.
Dan,Tv is very far from being on its "last legs'---but I realize that there is little point in trying to show you why. As for Netflix having no reason to share its usage data with anyone but its investors---absolutely right. I agree.Nevertheless, now, no matter how Netflix tries to hide info on its audience dynamics, the rest of the TV and media world will have an independent and objective assessment---no matter how badly Netflix feels about it.As for the question of accuracy, if Netflix wishes to prove Nielsen wrong, all it has to do is turn over its show by show data to an objective auditor who has access to Nielsen's rating estimates for the same content. Let's see how much they differ. Otherwise, its protests that Nielsen will get it all wrong are nonsense.
Having seen up close how nonsensical Neilsen data can be, I don't think expecting Nielsen to get delivery of Netflix data wrong is too far off the mark.
They were the only game in town for so long that they became complacent and haven't kept up with industry improvements. It is irresponsible to think that they can adequately deliver exact data from any market, so how could they adequately analyze Netflix?
adgirl, could you provide an example where the national--not the local market--- Nielsens were proven to be way off "the truth". I'm sure that the agency time buyers and the big TV ad sellers would want to know.
In a digital world, it was barely a year ago Nielsen started to phase out paper diaries.
That’s all that needs to be said.
Nielsen needs Netflix to try to remain relevant. Netflix doesn’t need Nielsen.
I'll bet you London to a brick that Netflix is using stream counts as audience, whereas Nielsen is using the universally accepted 'average minute audience' (as per MRC).
Stream counts (de-duplicated by IP address and device if possible) most closely approximates the reach of a TV programme and not the average minute audience.
And of course as other people have noted there is the issue of the playback window, and the number of playbacks as wel.
That is, both are 'right' in their own way. However, one is the accepted standard for talking about audiences as opposed to talking about cumulative audiences. One of the problems with accumulating audiences is that you can end up with an audience bigger than the population - does that sound familiar to anyone.
Of course Netflix is perfectly allowed to hold onto their numbers. And of course Nielsen is perfectly allowed to publish their average minute audience numbers as per MRC accreditation.
Dan, Nielsen has never used diaries as the basis for its national ratings. They were utilized until 1987 to provide audience comp estimates but set usage was always tracked by meters and this has been true since 1987 with the peoplemeter panel. When you refer to paper diaries, you mean those used in smaller TV markets for local rating reports because the stations in those ares are too poor or unwilling to pay more for meter ratings. As for Netflix, Nielsen will measure its audience with a 40,000 home meter panel containing 100,000 residents. If Netflix has the audience it claims to have this is a perfectly acceptable measurement system and not at all outmoded.
I informed my class about this story. The analogy I made was that Netflix providing studios and advertisers with data is similar to having my students grade their own midterm exams (that got a lot of laughs!).
Obviously, advertisers are extremely anxious to know how Netflix programs are performing because of the product placement in the shows--I'm surprised no one on this thread, or the author of this article, mentioned that.
Netflix is undoubtedly a great company and it has many insights to offer about how to use its own metrics. However, studios and the advertising community would be ill advised not to conduct their own analyses using 3rd party industry-sanctioned AUDITED data. The subsequent validated data sets would be enormously helpful to the studios and advertisers, so they can make sound business decisions. Netflix's library decreased by approximately 33%, so it's quite possible some independent evaluation has already begun that caused select studios and advertisers pull out.
If someone has a reasonable explanation for trying to discourage my clients from foregoing due-diligence by asking us to blindly trust another company during business negotiations without verifying results, I would be very interested to hear that explanation.
Great post Craig.
My only quibble is that 'auditing the data' is a monstrously huge task that would take forever and make the results so late that the data are rendered basically useless.
What is needed is to have the internal data collection, calculation and reporting system audited and accredited. As long as all data passes through that 'audit pipe' then fair is fair. From time-to-time unannouced spot checks can be independently conducted to ensure compliance to the accredited rules.
It's a bit like having to renew your driving licence every few years rather than every time you get behind the wheel of a car.
Netflix has no need of outside data. It can tell with perfectly clarify what all its subscribers are doing. I can watch episode three of Mindhunter up to the 7:08:23 mark and then abruptly switch over to episode 10 of season three of Orange is the New Black, starting at 6:03:12 where I left off last time. Now how is that possible unless Netflix is the one making it possible? It doesn't cache the 50 series I have in my queue (with multiple seasons) on my poor little Roku box. Netflix could follow me all around and then correlate my behavior to others like me, or completely unlike me, or whatever it likes. It has data that would make Nielsens head spin.
Nielsens is trying to approximate the data Netflix has in order to sell it to the companies that make or license shows for Netflix, to use in negotiations. I'm sure David Fincher would like to know exaclty how popular Mindhunter really is. Maybe he's getting screwed out of money by those Netflix cheapskates.
Netflix of course will deny that Nielsens' data is accurate. They have no reason to help people negotiate against them. And the data is bound to be off, because they're not tracking mobile or overseas viewing, the two biggest growth areas of streaming!
Netflix doesn't have to prove they have better data than Nielsens. They must have better data, or they couldn't serve up series, movies and episodes, and let us all pause, fast forward, rewind, etc, with such laser-specific control. That's all being tracked by Netflix so they can make it happen. How could Netflix not have that data?
And Netflix has no motive to validate Nielsens data, because Nielsens is selling it to people to use in negotiations against Netflix. So Netflix will just say, "sorry, no way off" regardless of whether the data is accurate. And will never provide numbers of their own becuase they don't have to.
All this will be good for is internet arguing when we all find out that Iron Fist was super popular anyway despite being generally hated. That will be fun, but the rest of it...pffft.
They are not foolish enough to destroy their brand with ads. Maybe in several years they might launch a lower-cost ad-based tier, but there will always be a premium, ad-free service.
Netflix doesn't even disclose series-specific ratings to investors! All investors get is overall subscriber numbers. But it doesn't matter to investors whether 109 million people are watching The OA or Adam Sandler or Ozark or whatever. The bottom line is what counts - and growth trends.
Why should Netflix provide data to advertisers when it doesn't have advertiers?
As for studios - either they are licensing content to Netflix or making it for them, and either way I'm sure they'd love to have some ammo to use in negotiations with Netflix to get more money. Which if successful, will be passed on to the subscriber.
That's the kicker, I wonder if the people participating in the Nielsens study know it can only be used in one way - to raise their Netflix rates? I know if Nielsens comes calling, I won't be interested in cooperating. Nobody should be.
When we say that information form a neutral outside party---Nielsen--about Netflix should be available we don't mean just to advertisers, we mean anyone and everyone who cares---the studios, the broadcastTV networks and basic cable channels as well as other SVOD content suppliers who compete with Netflix for viewers, financial analysts, the FCC, etc. Moreover, it's not always a question of the producer charging Netflix more for a show ---because of knowledge about the audience---and this being passed on to the subscriber. In many cases quite the opposite happens and Netflix shows bomb, which means that they are either cancelled or the producer is paid less for making a new series of installments.