No conversation about Nielsen in the media business is complete without the word “frustration.”
There’s little about the pre-eminent TV measurement company that hasn’t drawn complaints, particularly its slowness to track viewing across VOD and mobile devices. As NBC Broadcasting chairman Ted Harbert told a packed industry crowd at the Hollywood Radio Television Society luncheon earlier this month, “They say they’re on it, but everybody in this room knows that they’re a monopoly, and they’ll get to it when they get to it,” he said. “It’s always coming next year. And I’m actually hoping it comes next year.”
Harbert may finally get his wish. With the new fall TV season just under way, Nielsen is about to officially announce new details of a twofold initiative that may represent the biggest step forward the company has made toward tracking viewing across multiplying screens. It certainly won’t be enough to assuage all of the company’s many critics, but it should at least help ease the first of the two distinct crises facing the entertainment biz due to the rise of digital distribution.
The first problem is that there’s no satisfactory way to harmonize data about video content and advertising viewed on TV, online, mobile and other platforms. The second worry, which may prove just as vexing, is a business issue: Digital distributors like Netflix, Amazon and Google keep much of their usage data proprietary — they see it as a strategic advantage to hold that information close to the vest.
But in the near term, the pressure is on Nielsen. “Networks are starving for a number they can publish that really represents their audience not just on TV but across all platforms,” said Eric Solomon, Nielsen’s senior VP of global audience measurement. “I think it will start changing the narrative that ‘people are not watching TV shows.’ It’s that they’re watching on different platforms.”
Viacom Media Networks chief research officer Colleen Fahey Rush is taking a wait-and-see attitude toward Nielsen’s promised improvements.
“They have a lot to deliver,” she said, describing the measurement company as “kind of like the bad boyfriend who keeps letting you down but you can’t break up with him, because he’s the only boy in the whole school.”
This week, Nielsen is expected to publicly announce to clients a timetable for a long-awaited development: In September 2014, the firm is aiming to be able to attribute linear TV viewed on smartphones and tablets as part of its National TV ratings. The approach, the culmination of more than three years’ work, will for the first time provide a single, consistent measure of live programming viewing across both TV and digital, according to the firm.
That means live TV streamed through such apps as ABC’s WatchABC, as well as on iPad video apps from cable operators, can now be counted toward the total numbers on which programmers get paid by TV advertisers.
“What we’re trying to do is bring a lot of the TV model to the digital world,” Solomon said. “This will find Nielsen-coded content on any platform.”
In tandem, Nielsen is working to deliver Digital Program Ratings, which measure consumption of TV content online delivered without the national commercials (replaced with digitally inserted online ads), in the first quarter of next year. The firm announced a pilot program in April that includes participation from ABC, CBS, Fox, NBC, A&E, CW, Discovery, Univision and AOL.
The Digital Program Ratings data uses the same methodology as Nielsen’s Online Campaign Ratings service, which is based on about half of the homes in its National People Meter panel (about 10,000 households that also have agreed to let the company monitor their online consumption). To glean age and gender data, and cross-reference those with the viewing numbers, Nielsen cut a pact with Facebook, which has about 179 million monthly active users in the U.S.
The two Nielsen digital initiatives support both linear TV and digital ad models, according to Solomon. “A key learning (point) for us was that no two clients are approaching this question of distribution in the same way — they need flexibility to choose the monetization model they use,” he said.
Do the Nielsen efforts go far enough? Industry observers are encouraged by the projects, calling them a step in the right direction, but said they’re not a panacea. “The fact that (incorporating mobile viewing into TV ratings) is still a year of shows we’re not moving at the speed that marketers and consumers are,” said David Cohen, chief media officer for ad agency Universal McCann.
About the grumbling that Nielsen has taken far too long to progress in this area, Solomon said the company can’t unilaterally alter its national TV ratings given the billions of dollars in advertising that’s sold based on the data. “Any change to that is thought through and analyzed to death, and that’s one of the reasons why this seems to take so long,” he said.
How big was the total audience for an individual episode or an entire season run, as seen on television, DVR, video-on-demand, websites, smartphones, tablets and other devices? And what’s the composition of that audience? Today what’s available are, at best, educated guesses. TV programmers have taken the lead in trying to solve this puzzle to support their advertising goals. Now, Nielsen and other research firms may be on the cusp of delivering useful metrics that content companies and marketers have been impatiently waiting to get their hands on.
This has been an elusive pipe dream in the TV biz for years. Networks need a consistent measure of unduplicated reach and frequency across all screens — something that the standard bearer for TV ratings, Nielsen, has been unable to deliver. Programmers also want data-driven insights so they can fan out their content based on what works on each platform, given that a growing segment of their audience is migrating to mobile and Internet viewing.
At a higher level, they need metrics to show that TV, with traditional ratings eroding, is not dying: It’s only fragmenting across distribution points.
“My company is evaluated by Wall Street on our (television) ratings, and that’s an increasingly poor way to understand our performance,” said Fahey Rush.
Why is measuring content consumption across disparate touch points so hard? To date no single entity, neither Nielsen nor anyone else, can compile the information across a single, consistent source. Melding different panel- and census-based data sets together — into what industry execs term “Frankenmetrics” — is fraught with complexity because that information is collected based on different methodologies.
“On one hand, I feel like we get more and better data from a variety of channels we didn’t used to have,” said Tim Van Hoof, assistant VP of marketing communications for State Farm, which spent $778 million on advertising in 2012, according to research service SNL Insurance. “Where we are falling short is, how do you weave through the volume and variety of data sources we get and aggregate them in a sensible way?”
In the absence of action by Nielsen, several programmers have placed bets on other research initiatives.
Perhaps the most ambitious is ESPN’s hybrid Project Blueprint, which the sports cabler began testing in February with a mix of data derived from cable set-top boxes, comScore and Arbitron. “It is a layer of quantitative data that the industry desperately needs to understand net reach across platforms over an extended period of time,” said Artie Bulgrin, ESPN’s senior vice president of research and analytics.
ESPN plans to present a full year of usage data showing the network’s cross-platform programming reach at its 2014 upfront next spring.
Nobody aside from ESPN, however, is using Project Blueprint today, and the methodology still awaits industry vetting. (Meanwhile, Nielsen is in the process of acquiring Arbitron for $1.3 billion; under an agreement with the Federal Trade Commission announced last Friday, the company must continue to support the Project Blueprint initiative.)
Today, no other cross-platform research is as far along as Project Blueprint, said Jane Clarke, managing director of Coalition for Innovative Media Measurement. CIMM, whose two dozen members include media congloms, ad agencies and advertisers, was formed in 2009 to promote new research approaches in response to what the founders perceived as Nielsen’s lethargy in the area.
“People have been complaining about this for years, ever since we started going cross-platform,” Clarke said. “I think we’re starting to see the potential with Project Blueprint.”
A separate data shortfall for Hollywood has arisen with the emergence of tech players like Netflix, Google and Amazon as significant content distributors.
Third-party estimates exist for electronic sell-through services such as Apple’s iTunes; that’s because individual suppliers know exactly how many titles they move through EST partners in any given period. But no such data is available for streaming services, which are a black hole for the biz.
Exhibit A is Netflix, which refuses to release information about what its 30 million U.S. subscribers watch in any detail, either publicly or privately, to content suppliers. For competitive reasons, the company doesn’t even cite how many titles it offers through its streaming service.
Netflix says it does not provide usage figures because it isn’t an ad-supported service — unlike traditional TV networks, it doesn’t have to tell an advertiser how many people tuned in to, say, an episode of “House of Cards.”
Company execs defend the policy by saying divulging metrics would establish a negative public perception about the content it carries, a downside that it, as a subscription service, doesn’t need to expose itself to. Releasing viewing numbers for Netflix originals, for example, would create “performance pressure around these shows, which is very unnecessary,” chief content officer Ted Sarandos said in a recent interview with British newspaper the Guardian.
And with Netflix’s expansion into original series, the data vacuum also lets the newcomer build buzz it might not enjoy if it reported audience size.
“People get carried away with these perceptions that (Netflix’s) ‘Orange Is the New Black’ is the new must-see show,” said Will Somers, head of network research for Fox Broadcasting.
The fact that the cloak of secrecy extends even to the content companies that license their products to Netflix makes it difficult for studios to figure out how valuable its deals with Netflix are relative to other distributors.
Netflix’s “opaqueness about the consumption data was very unfamiliar to me,” said a top media research executive who requested anonymity because the exec’s company still has a business relationship with the Internet streaming company. By keeping the numbers vague, “They have all the leverage in negotiating the next deal.”
FX Networks CEO John Landgraf went on record on the subject at the HRTS event a few weeks ago. “Part of Netflix’s strategy is not only not to release that information to the public but not to tell the content producers and providers what the value of their content is on Netflix’s system,” he said. “And that’s a huge disadvantage to sellers.”
Netflix declined to comment for this story, referring to Sarandos’ previous remarks.
Meanwhile, Google also keeps its big data to itself, preferring to steer businesses toward Google Analytics for information about the Internet giant’s services. That extends to YouTube, the Internet’s No. 1 video site by a long shot, which serves more than 17 billion views monthly in the U.S. alone, according to comScore. Services like comScore tabulate aggregate data for websites, but don’t have the granularity to report on the performance of individual titles.
“If I run a clip of ‘Conan’ on YouTube, Nielsen can’t count it, because Google will not share their data with Nielsen,” said Jack Wakshlag, chief research officer for Turner cablers.
If Internet publishers pooled data, it would provide much greater transparency for the industry, according to Wakshlag. “I would know what my share of the business is, and I can figure out if you underspent with me,” he explained.
Added Seth Demsey, AOL’s senior VP of advertising product and strategy: “Everybody has their own proprietary data they’re sitting on. It’s a little disheartening. If we share across industries in more productive ways, we could meet client objectives better.”
Google, in a statement, said it’s continuing to explore new measurement options without explicitly committing to sharing data externally. “We know buyers want meaningful measurement, which is why we’re investing broadly in brand-friendly metrics,” the company said.
In reality, there will never be a perfect way to measure content consumption across all distribution points. “I don’t think there’s any such thing as a silver bullet,” said Nielsen’s Solomon.
Jon Mandel, CEO of PrecisionDemand and a longtime ad industry exec, has a vested interest in attacking the status quo. His company analyzes set-top box data, with the promise of letting marketers match up ads aimed at specific consumer segments with TV shows — correlated with purchasing data. It’s an approach that’s far more accurate than they can achieve using Nielsen’s age and gender data.
From 2006 to 2009, Mandel worked for Nielsen as CEO of NielsenConnect, a now-defunct division whose mission was similarly intended to connect ad buys with consumer buying behavior.
“The problem is, the current (Nielsen) system works well enough for the way the business has run,” he said. “But one day they are going to wake up and find they don’t have a business.”