New WBD Report Makes The Case For Getting The Measurement Basics Right
 |
By Alyssa Boyle
|
Getting gobbled up by Paramount (or, rather, Skydance) isn’t the only thing on Warner Bros. Discovery’s mind.
It’s also thinking about ad measurement.
Connected TV media buyers and sellers continue to lament the lack of standards in ad measurement. But a new white paper from WBD has some key takeaways that the company hopes can help point the industry in the right direction.
Throughout last year, WBD investigated the data and methodologies of five top video measurement providers: VideoAmp, iSpot, Comscore, Innovid and Samba. (If you’re wondering where Nielsen is, the TV ratings titan doesn’t allow side-by-side comparisons with competitors, sources have told AdExchanger in the past.)
WBD conducted a side-by-side comparison of these five providers in five key areas: universal audience estimates, impression and viewability counting, cross-platform coverage, identity solutions and personification. The report covered more than 50 campaigns across WBD’s traditional and streaming TV. WBD also commissioned a survey of 154 agency and marketer decision-makers from Advertiser Perceptions. Then, WBD turned the findings into a white paper published during Advertising Week New York, titled “The Unified Reach Paradox.”
The most telling part of the white paper is the fact that the industry is still struggling with measurement, said Mariel Estrada, head of video currency at Omnicom Media Group, one of the agencies that provided feedback to WBD about the benefits of the research.
WBD wasn’t looking to score measurement providers or rehash the same conversations the industry has been having for many years. The objective is to provide more information that can help set standards and expectations about how video ad measurement should work, said David Porter, head of advertising research, data and insights at WBD.
More data, please
So, what did that white paper reveal? In a nutshell: audience reach measurement across platforms “is not very mature at all,” Porter said.
That lack of measurement maturity stems in part from the degree of variance across providers.
“We knew the results would not be consistent,” Porter said, referring to WBD’s research. “But we didn’t expect the variation to be as wide as it is,” he said, calling the degree of difference across measurement providers “quite alarming.”
Each measurement provider must have competitive differentiators to survive as a business, Porter said. But the degree of difference in results across providers far surpasses a rounding error, he said. The report found several instances of one vendor’s estimates nearly doubling that of another.
One example from the report is a cross-platform campaign for a financial brand in Q4 last year. Comscore counted 43.5 million impressions among people ages 25-54, whereas iSpot counted only 24.6 million cross-platform impressions for the same campaign and demo.
It’s quite a difference and explains why nearly three-quarters (73%) of the agencies and marketers WBD surveyed named “unified reach” among the most important missing data points in their measurement equations.
There are a few reasons for the variance. One reason is the difficulty obtaining necessary data from TV and streaming platforms themselves. Platforms vary in the type and granularity of viewership data they’re willing to cough up because “that data is precious” and a competitive advantage, Estrada said. When platforms do hand over that data, they typically share more directly with an agency client than they would with a third-party measurement provider, she said. This lack of transparency means agencies are left unable to create a comprehensive picture of viewership across media companies.
Now what?
Cleaning up measurement requires an honest assessment of where measurement vendors can differ and where they need to reach consensus.
According to Porter, it’s acceptable and healthy for measurement vendors to differ in their identity solutions and personification methodologies because they inform competitive differentiators, such as co-viewing. For example, some measurement platforms offer co-viewing measurement features to help marketers understand more about who within a household might be watching something.
But there are some standards that all platforms should adhere to – and instances where disclosures would be beneficial, Porter said.
First and foremost, counting impressions and viewability should be standardized, he said.
“It really is kind of crazy that we as an industry are still debating what constitutes a video impression,” Porter said. He didn’t name names, but we can: YouTube took a stab at creating its own measurement standards a few years ago and pissed off other publishers. And over the summer, it switched up what counts as a video view on its platform. Now, YouTube counts views from the moment a video play begins. (Just to give one example.)
Universal audience estimates should also be consistent across measurement providers in an ideal world, Porter said. Examples of data sources include census data and the Advertising Research Foundation’s DASH TV Universe Study. And if measurement providers choose different data mixes to create a picture of reachable households, then those choices should be disclosed, Porter said.
Porter expressed a similar sentiment for platform coverage.
“In no case does every single [measurement] provider cover every platform out there,” Porter said, acknowledging that the long tail of TV content out there “gets very long.” But those gaps in viewership coverage lead to projections and modeling technology that isn’t always accurate. In those cases, Porter said measurement vendors should at least disclose how and when they use projection modeling to estimate what viewership looks like across platforms. Vendors should try the “nutrition label” approach to tell clients what “ingredients” feed their projection methodologies, Porter said.
Ultimately, Porter said, “standards and disclosures are the path forward.”
Reassessing outcomes
Of course, no assessment of TV measurement nowadays would be complete without acknowledging outcomes.
“Everybody has been talking about outcomes” because that’s where the marketplace is going, Estrada said. But at the end of the day, she said, reach and frequency is a measurement baseline that the industry cannot afford to overlook.
According to that buy-side survey in WBD’s white paper, 79% of respondents said they can’t fully trust campaign outcomes without confidence in unified reach.
The TV and streaming industry is a bit too hyper-focused on business outcomes, Porter said, naming conversion APIs (CAPIs) as one example of how today’s infatuation with outcomes materializes. “CAPIs focus on measuring bottom-of-funnel outcomes,” Porter said. But what the industry needs to do is “broaden the definition of outcomes,” which start with reach, he said. For marketers, “changing someone’s perception of your brand is an outcome,” for example, he said.
Growing brand awareness and affinity is how a marketer keeps their brand growing and thriving in the long term. Which is why TV and streaming is increasingly referred to as a “full-funnel” channel, he added.
In short, WBD’s white paper “should really just be the beginning of the conversation,” Porter said. “It should spark dialogue and conversation about what we do next and how to make better decisions” as an industry.
In other words, bring in the next phase of TV ad measurement.
|
