GroupM is weeks away from finalizing what data its clients use for their upfront advertising buys, but it is likely at least some GroupM clients will be ready to make a switch from Nielsen's legacy ratings data in time for the 2023-24 season, according to the media-buying giant's head of research and and investment.
"We have some clients who really want to move. There are some who are interested, but want our point of view on it," GroupM Executive Vice President Bharad Ramesh said during this week's episode of the agency's business intelligence team's "This Week, Next Week" podcast, which focuses mainly on the potential role of a JIC -- or a joint industry committee -- for setting a new market currency in the U.S.
The discussion, which was prompted by last week's announcement by the U.S.' biggest TV network owners that they has already formed one to do just that, was described as a premature statement by Ramesh and GroupM Global Director of Audiences Research Simon Thomas, who noted he has had 40 years experience working with various JICs around the world, and characterized the U.S. networks' initiative as more of a MOC, or media owners committee than a genuine JIC, which is "tripartite" and requires equal representation from advertisers and agencies as part of its decision-making process.
Semantics aside, Ramesh and Thomas indicated the U.S. networks initiative is a step in that direction and implied the U.S. is on the verge of a massive overhaul of its market currency that will transform the way advertising is bought and sold, and how its effectiveness for advertisers is measured and analyzed.
"I would say that in three years by 2027– maybe four – it's going to be all about audiences," Ramesh predicted, noting that historical measures like Nielsen's based on a so-called "opportunity to see" an ad bought by an advertiser, would be replaced by measures of actual exposure, and potentially even higher orders such as actual "attention" paid to them.
In the near-term, the GroupM experts implied the U.S. marketplace will become increasingly complicated by an array of alternative currency vendors pitching to be used as the primary or secondary basis for advertisers TV, as well as cross-media, advertising buys, which could create marketplace confusion.
“Potentially, people are worried about that happening within the total video marketplace in the U.S, where we end up with a VideoAmp metric, an iSpot metric, a Nielsen metric, a Kantar metric, and so on," Ramesh cautioned, adding, while a U.S. JIC ultimately could lead to a "standard industry metric," the other currencies could be used as the basis of negotiations for some advertisers and networks.
"If you want to trade on top of that, you could do that. It's not a problem," he said.
While a number of big TV network owners, including NBCUniversal, Warner Brothers Discovery, Paramount and Fox, already have begun "certifying" their own alternative currency suppliers, including iSpot.tv and VideoAmp, Ramesh said it's too soon for many clients to make a shift from Nielsen's legacy ratings in time for this year's upfront, and that -- in retrospect -- an industry-wide initiative should have begun in September 2021, "the moment Nielsen lost its accreditation" from the Media Rating Council (MRC).
Perhaps the most significant issue raised by Ramesh and Thomas was how an eventual transition to true exposure-based audience measurement will impact how much advertising U.S. advertisers buy and what they pay for it, because extraneous "frequency" will be removed from most advertising buys.
"The total budget the clients spend is probably going to go down," Ramesh warned, adding, "because the fact of frequency goes away."
In other words, he said, the potential of "unified" exposure-based research across video platforms to remove extraneous frequency form the mix, "for 10% of 20% less budget, maybe my cost per point was higher... but I'm actually in the end spending less money to hit my objectives. That is a very, very real thing."
Ramesh said the costs of "non-working" media would likely go up for advertisers in that scenario, because they'd be paying more to process the data, but ultimately there would be a net savings on the total cost of buying advertising.
"That's what we're looking to see in the next three months -- when you look at all the data coming through from unified measurement -- is, 'Hey, what is the actual, true, reach and frequency that we are getting.'," he explained.
He added that the tension ultimately would come from pushback from the "sell-side," and GroupM's media buyers, to negotiate how they "protect your CPM basis" in a marketplace where advertisers buy fewer overall ads.
"That's the pressure. The data wants us to spend less, but for the network and the trading side, they will not tell you to cut the budgets," he said, adding, "That is the next shift that's going to be the tug-of-war. If not in this upfront, definitely by 2024-25."
What, if any role, a JIC ultimately plays in any of this will depend on a number of variables, including complexity, potential legal haggles, and the fact that JICs historically have been slow to move, according to GroupM's Thomas.
Joe, people keep confusing the terms "currency" with who is supplying it. What the basic "currrency" should be is obvious. It should be a valid "audience" measurement and if by "audience"we mean who watched a particular segment of program content or a commercial, we have never had a valid "currency" for TV. Why? Because nobody felt the need for one---it was always assumed---or hoped---that people who self-defined themselves as "program viewers" almost never left the room and almost always were attentive---no matter what was on their screens. Yet numerous studies kept showing that people often left the room and or were paying no attention and that this was especially true when commercials were on their screens. We cover all of this in our upcoming report, "Total TV Dimensions 2023".
The only way to get a true "audience currency"for national TV, including streaming, with enough of a sample to permit show by show data even for low rated programs is to use a large enough panel that represents all TV users, not just ACR set usage, and includes an "eyes-on-screen" determination so, for the first time, a buyer knows how many and who watched a commercial. All of the other "metrics" may or may not have value as add-on indictors for those buyers who find such info useful. Yet, we are not heading in this direction. Instead we are going to mix and match "big data" set usge findings with exactly what we have utilized since the early 1960s---claimed program "viewing" stats derived from a relatively small panel.
@Ed Papazian: Not sure about the word "confusion," so much as different constiuencies having different interpretations of what the word "currency" means. My own definition, developed over years of covering all sides of the business, is that it is what two sides of a marketplace -- buyers and sellers -- agree do do business based on.
But that gets funny when people start augmenting the term with things like "alternate currencies" (if it's a currency, it's not an alternate one) or "secondary currencies," etc.
I think you're describing something else, like what in your view, from a methodological point-of-view, should be the standard bearer of TV audience measurement.
Ed, I agree that TV includes people who maybe vaguely watch the TV. I did a study in the early 2000s and it showed that during ad breaks there was a clear dip in the audience. So, it does show that panelists do log themsleves out and then back in again, but I don't think that is not ubiqitous so I am sure that the rating is on the high side. FYI, the average was just under 5% and peaked around 8%.
But I think that the measurement 'killer' is those who remain in front of the TV, don't log out and ignore the ads (N.B. not necessarily the broadcaster's fault).
There is software and hardware that can do studies on presence, focus and attention. But I can't see that as a continuous measurement as being feasible. One-offs (say, over one week) can be done and can be used as an indicator. If we knew that the average attention loss in prime-time movies by W25-54 was 28% I reckon we could use that to differentiate.
I think we could do limited (in sample size and duration) attention research studies and apply them as a 'attentive viewing factor' to the count of bums on seat (aka ratings) and have a pretty good 'attentive audience' estimate that would probably meet commercial needs.
I would also like to similar rigour in the digital world, rather than the virtually ubiquitous reliance on one device talking to another device as being gospel. It's improving but still has a way to go.
John, there are fairly meaningful differences in both presence in the room and eyes-on-screen measurements by program and certainly for individual commercials. Long ago at BBDO we developed norms---or averages--- of the sort you suggest but then found that we were still off by about 25-35% in terms of specific shows, dayparts and demos---age and education.
What I'm proposing is a national panel along the TVision model---with its members opting in plus all sets in the home measured for attentiveness via their webcam system. As the average attentiveness level for commercials is about a third of the "audience" you need at a minimum about 75,000 homes to match what Nielsen has been giving us---but I would prefer double that number due to ever fragmenting ratings. The findings would be appplied against a much larger base that combines ACR set panels with STB panels---the latter involving measuremnet of non-ACR sets as well as the "smart" ones.
This isn't going to be easy as there's a lot of statistical weighting involved in melding the two "big data" panel bases into something that approximates a nationally projected panel for TV set usage. But if such a melding can pass the MRC's sniff tests---which I think is possible----then the viewer-per-set factors from the attentiveness panel can be projected against the much larger set usage base to yield relativeley stable data.
Of course, such a system will cost more but so what? We are talking about a medium which attracts about $100 billion in annual ad revenues---"linear" plus streaming---to say nothing about other incomes. And its TV, which most branding advertisers are wedded to. So it's long past time we began to get it right.
To Joe's point, it's either a currency or its not! "Alternative currencies" reminds me of the now infamous, "alternative facts" comment regarding the 2017 US inauguration attendance.
To Ramesh's point, the current auxilliary measurement services offer invaluable enhancements for planning and buying but over and above a fundmantal consensus TV/video/streaming currency.
Per Ed, the currency needs to be persons based Eyes-On screen. Of note, Nielsen is I believe the JIC measurement provider in some markets and provides quality ratings albeit the currency is controlled, managed and priced by the JIC on behalf of the industry.
Disagree with Simon. Respectfully suggest that JIC's have surely proven themselves to be the "least worst" currency providers at a minimum based on across all busienss dimensions. The sole US JIC, GeoPath, (formerly TAB), has saved the industry $millions; lead the entire media industry with its Eyes-On or Visibility Adjusted Contacts measurement which stopped the discounting in OOH; and has continuously improved its complex integrated hybrid measurement approach. BARB in the UK is also another example of the overall value of JICs.
Last and not least, advertisers and their media agency partners are poorly served by any media currency managed and controlled by the sellers whether "certified" or not.