A study commissioned by the Coalition for Innovative Media Measurement (CIMM) over the summer suggests the TV industry still can’t agree on the point of panels.
In fact, TV measurement panels are so controversial, I moderated a panel about them this week. (Ha, get it?)
But the debate swirling around panels has never been about getting rid of them completely, Helen Katz, EVP of research at Publicis Media, told me on stage. “[What] we’re still debating [is] how to build and use them correctly,” she said, which includes pairing panels with other types of data to ensure measurement is as accurate as possible.
In the meantime, panels are causing a bunch of confusion.
Calibration, what?
Companies and trade orgs are all creating their own different viewership panels, which they’ve taken to calling “calibration panels” to highlight certain very technical use cases.
The term “calibration” doesn’t have a consistent definition and seems to change depending on who you’re talking to, says Joan FitzGerald, CEO of Data ImpacX, who helped conduct the study.
“Panels are still the standard for TV measurement,” FitzGerald said. But instead of the audience panels Nielsen gets dinged for on the daily, calibration panels are in right now.
And despite the super nerdy name, these panels have a fairly simple function, which is to verify the accuracy of bigger data sets, including viewership and census-level data, that are becoming a more popular foundation for measurement due to their specificity.
Examples of calibration panels include those of smart TV companies Vizio and Samsung, which created their own respective panels over the past year using viewer data and automatic content recognition (ACR).
Trade orgs, such as the VAB and ANA, are also working on their own panels, although their approaches also differ. The VAB is using ACR plus other viewing data from another panel provider (think Kantar or HyphaMetrics as a hypothetical), while the ANA is building an ID it can integrate with alternative measurement providers (like, hypothetically, VideoAmp or iSpot) to identify TV audiences.
Despite the glaring lack of consistency between them, however, these panels have a shared objective – namely, to fact-check data in order to support better measurement and, therefore, more effective targeting.
Measurement companies, for example, combine panels and big data to reveal more information about TV households, such as age, gender and ethnicity. The additional information helps flag underexposed audiences advertisers should be reaching and also corrects errors in reporting (such as a non-Hispanic household watching telenovelas – possible, but not as likely).
The ability to identify individuals within a home is also why advertisers turn to panels for help measuring co-viewing, which can be useful for more targeted campaigns, Publicis Media’s Katz says. Even so, there’s still more value in targeting households (as opposed to individuals) when the message of an ad is more general, she added.
On the same page
The bottom line is this: Panels are a resource for measurement and don’t directly determine the value of media buys as currency.
But that doesn’t mean they shouldn’t be standardized.
Both Media Rating Council (MRC) accreditation and the certification process that the broadcaster joint industry committee (JIC) is building are “critical” for ensuring buyers and sellers can agree on measurement that has “statistical reliability,” Katz says.
Data needs to be consistent enough to help plan and forecast future campaigns, and it also needs to flow through the planning and activation systems that buyers are already using.
Although compared with JIC certification, MRC accreditation is a “much more rigorous process” that takes a very long time, Katz adds.
So long, in fact, that the question of whether MRC accreditation is necessary is a subject of debate within the ad industry because time is of the essence for the adoption of new measurement currencies.
In the meantime, says Carmela Fournier, VP and GM of data at Comcast Advertising, “the JIC is setting the bar for the types of standards needed to have good, accurate measurement that’s acceptable to both sides.”
Not sure Nielsen would agree with that, though.
Are you enjoying this newsletter? Let me know what you think. Hit me up at [email protected].
