Another fall television season has arrived, bringing with it the usual assortment of new series and returning favorites. But this particular fall season kicked off with a dramatic plot twist: in September, the Media Rating Council (MRC), a nonprofit that oversees accreditation for media research and rating purposes, suspended support for Nielsen Media Research’s national and local TV ratings services, for decades the measuring stick used to determine programming success and establish advertising rates.
Nielsen is working to remediate issues raised by the MRC, and broadcasters and advertisers are sticking with its ratings system to conduct their business, even in the absence of the MRC’s seal of approval. But Nielsen’s struggles have brought new attention and awareness to how much legacy ratings models are stifling the media industry’s progress, and amplified calls for a more scientific, innovative approach to audience measurement.
“We need a shift in how we measure and how we monetize, and the legacy of Nielsen and the legacy of media-mix modeling has prevented that,” an anonymous media buyer told Adweek. “There is so much data and technology today available in everything we do. I would hope this would provide an opportunity to completely change, shift, recreate the way we evaluate, measure, and monetize.”
Television needs a new and improved ratings system — one that is more representative of the audiences networks and advertisers are trying to reach, and delivers more accurate data.
Broadcasters and advertisers have been frustrated with Nielsen’s ratings system for years, partly due to market dynamics (specifically, Nielsen’s stranglehold over legacy media like TV and radio) and partly due to methodological concerns. This drumbeat has grown louder as the media industry has become more accustomed to activating through digital channels offering superior targeting and attribution capabilities.
Nielsen ratings are calculated based on a sample of 40,000 homes and 100,000 people the company says are demographically representative of the American population as a whole — numbers that critics say account for too small a fraction of the estimated 121 million homes with TV service. Nielsen has also come under fire for underreporting young adults and failing to adequately address out-of-home TV measurement.
The relationship between the networks and Nielsen reached a tipping point in early 2021 when the Video Advertising Bureau, a trade group that promotes broadcast and cable networks, alleged Nielsen significantly undercounted TV audience numbers and usage during the COVID-19 pandemic, negatively impacting broadcast ratings that are already suffering as viewers migrate to streaming outlets (see graphic below).
Speaking this August on Discovery Inc.’s second-quarter 2021 earnings call, President and CEO David Zaslav called for “better data” than what Nielsen delivers.
“You’re dealing with a very antiquated delivery system. We’ve all learned how to get along with it. We do it by augmenting with our own data,” Zaslav said. “But recently, [Nielsen’s] just been wrong. It’s one thing if you have an antiquated system and then you augment it. But the antiquated system itself is unreliable. And so, as an industry, we have to figure out how to deal with it. We’re competing with the likes of Google and Facebook, where they have the best data, the cleanest data. You compare that with this antiquated system, so we continue to work on our own. I don’t have a lot of hope for Nielsen. Somehow, as an industry, we’re just going to have to work our way out of it, from a technology perspective, and leave them in the dust.”
How would a post-Nielsen television ratings system differ from its predecessor, and how would it deliver the “better data” Zaslav craves? The future begins by focusing on the technical capabilities for ratings measurement.
Any viable alternative to Nielsen must combine larger sample sizes and more granular data with modeling techniques and audience panels that more legitimately reflect the overall viewing population — an evolution dependent on two key factors:
The most effective approach to an overhauled television ratings system would make use of this streaming data to offer networks and advertisers conclusive evidence of the number of viewers consuming a program at any given time — information bolstered by national surveys to make certain the audience represented by this streaming data accurately reflects the general population’s makeup and viewing behavior.
Why streaming data? Think about Google and Facebook: not only do both offer digital advertisers access to far larger audiences than Nielsen, but it’s also a lot easier to accurately determine when someone is scrolling through a web page or viewing a banner ad than it is to identify whether someone is watching TV in a bar or listening to radio in a public place. Moreover, advertisers can leverage any number of first-party data sources to tie the offline identities of the people and audiences consuming digital media to social media consumption or propensity scores — not to mention that Google and Facebook also support more precise behavioral traits for audience targeting than traditional media channels.
Another critical piece missing from legacy TV ratings systems is the ability to quantify marketing ROI. We’ve traditionally relied on media mix models to understand the impact of linear TV spots, but when you field an ad through digital channels, we can get a more complete picture. Those digital channels offer clear data on who was exposed and who wasn’t, enabling a deeper understanding of ROI through controlled tests or synthetic controls, if you have the tools to interpret it.
These are precisely the kinds of improvements and innovations the 21st century entertainment landscape demands. It’s a new era of television, and that calls for a new approach to measurement and monetization: as cord-cutting and multi-screen viewership continue to fragment traditional broadcast paradigms, it is becoming increasingly critical to know who is consuming TV content, as well as where and when — insights that legacy ratings systems don’t deliver.