Commentary
What If The Multiple TV Currencies Used This Upfront Clash?
- by Dave Morgan , Featured Contributor, April 13, 2023
There has been a lot of talk over the past year about the future of measurement and currency in the TV ad market. The “alt measurement movement” has been a hot topic in the ad trades, on stages at industry events, and over cocktails and meals wherever industry peeps get together these days.
A number of companies, both on the sell side and buy side of TV media, have made announcements they are prepared to transact on multiple rating currencies this upfront, which will certainly be an enormous step if it happens. But someone raised two very simple questions to me recently relative to the currency issue that I haven’t seen addressed in recent news stories or on the conference stages.
What do we do when the currencies don’t agree with each other? For example, one of the currencies could say that a show delivers 20% more viewers under 40 years old than another measurement service. The second question: Who gets to pick which one is used? Is it the seller, who’s likely to pick the one with the higher number and get paid 20% for a young demographic sale? Or the buyer, who is likely to want the one with the lower number, and pay 20% less on the buy?
advertisement
advertisement
Think this won’t happen? Of course it will. Just go back and look how things worked decades ago, when Arbitron and Nielsen battled in local markets with competing TV ratings (and radio too, by the way). It was an enormous issue that wasn’t resolved until Arbitron finally withdrew from the TV ratings market (and was eventually bought by Nielsen).
For sure, we’re going to hear a lot about competing data collection methodologies, measurement sample sizes, sample biases, balancing models, statistical smoothing, etc. That will be fun. But that won’t answer the question of what happens when real money is at stake. I can’t imagine that the advertisers -- the ones ultimately funding the market -- are going to eagerly step up and pay a bunch more money for a spot based on one rating if they have a legitimate basis to choose a competing rating with a lower number and a lower price for the media.
What do you think?
A juicy topic, for sure Dave. First, there probably needs to be more public transparency which reveals audience estimate differences across alt currencies. Are they systematic, i.e., are Alt currency A estimates always higher than Alt currency B's and C's? Or do the differences vary by program and/or daypart? How are media agencies able to know if they don't subscribe to all of the options out there? While the alt currency movement is a big step in the right direction, some standardization is needed to evolve alt currency. But I don't see that happening in the near term since alt currency opacity may be considered a marketplace edge for sellers and buyers alike, especially with advanced targeting. I suppose the two sides will have to sort things out, deal by deal, for the immediate future.
The currencies are sure to clash when mixing probabalistic panel data and ostensibly deterministic metered data. Among the features of machine-driven "big data" is the propensity to over count the big and undercount the small (hence gains previously seen in large demo breaks and declines in smaller demo breaks). Shortly after Nielsen lost its accredidation in 2021, I wrote a post about multiple currencies being the outcome of a process that had already been underway but accelerated by the MRC's move. About that loss of accredidation I wrote
"none of it may matter if large advertisers conclude that having a one media currency system requires more friction than extending the analytics they are already doing to arrive at a common denominator metric across the multiple media they are using. Basically, what has been the usual and sometimes accurate, if not always precise, way of validating media across multiple channels and types will continue to be disparate and over time become more precise. Advertisers will still want a common denominator metric against which to normalize their means of valuation across media channels. A lot of work has been done across more than a few marketing service providers and their clients to put pieces together that do not naturally cohere in order to get a comprehensive understanding of media performance. It's what's been done for ages in one way or another, and with more media connected to an IP address -- video (TV), audio (radio), digital out of home -- the data to do these analyses will come faster and be easier to centralize and process."
The flip side -- or maybe just reticulated facet -- could be greater consolidation of spending across a single media owner's universe of O&Os. I suspect this is what the largest legacy media groups are hoping for with their moves to shore up their own measurement systems. A major advertiser can put budget towards, say, NBCUniversal, Disney, and/or Warner and worry only about reconciling the small handful of currencies, should buys be placed with all of them, per the scenario above. Whichever way things go, it'll likely be messy.
Dave and guys, most if not all of the so-called "alternative currencies' are not currencies at all, nor are they what would be termed "audience" measurements. Rather, most fall into the broader and somewhat nebulus category of "qualitative" metrics--like attentiveness, pupil dilations, brainwave patterns, etc. some are mainly about website visits while others use big data panels ----basd on set usage not viewing---to profile household "audiences"--- presumably by demographcs--- to see how well one show "targets" desirable consumers relative to another show. And,yes, Dave, they are bound to show different results for different sellers and shows. But it wont matter as virtually all buys will be based on "impression" guarantees ( GRPs ) with ratings from the same source---the winner of the national "TV" rating contest---which, I assume, will be Nielsen's big data plus people meter design. In short, the sellers do not envision many competing "audience" sources---only one ----as without that there would, indeed, be chaos. And they have, in effect, said so.
As for the add-on "currencies" the sellers have also been very clear about their intentions. None of these will be standardized. Each seller can use whatever add-on "metrics" they prefer whenever they wish and can find a willing buyer. That's it, plain and simple. And, needless to say, when one or more of these add-ons is included in a buy the seller must pay the data supplier and such extra costs will certainly be passed on to the advertiser via higher CPMs. So it will be up to the buyers in each case to determine if the added information and any guarantees based on same are worth the extra price. Sometimes the answer may be yes, sometimes, no.
If I was an advertiser and required to rely on 'alt-currencies' I think I would pay for my advertising with 'alt-dollars'.
Dave: You really know the optimal, albeit can never be perfect, answer.
Establish a real JIC. This will ensure industry ownership of the data and its copyright for a single credible objective trading (basic truth-set) currency that agencies and sellers can augment as they always have with any media currency. It will save the industry $millions every year for years. Also be sure to include Nielsen in the stringently rigorous mullti-partite RFP research vendor process.
As you are aware some of us actually understand the term JIC and its parameters based on global experiences acorss many media!
100%! If the data sources underneath each measurement provider disagree, then the outcome is that each currency will literally indicate a different value.
NEWS FLASH: We see discrepancies between providers of up to 40% on identity and 60% on demo assignment. These data sets were built to improve targeting efficiency, not to provide precise, projectable data for media measurement.
The US market wants competition and diversity in measurement providers, but the answer isn't to introduce even MORE friction and confusion into the market. The industry needs to institute new standards for data accuracy for both identity and demography so that everyone can agree on the definition of foundational building blocks like identity and demos, and then differentiate in other ways such as context, attention, ersonification, content, and outcomes.
Scott, you are correct--if the so-called "alternative currencies" were really attempting to measure the audience. But they aren't. Most, if not all are purporting to measure some type of"outcome"---such as favorable brainwave readings or pupil dilation indicators, higher ad attentiveness, click throughs to the advertiser's website, short term increases in brand share of market, etc. or some way to determine which TV shows best target the brand's prime prospects. In theory, were all of these used by the same sellers ---for the same brand---and its ad messages were more favorably received via one network's viewers relative to another's, they should all indicate this positive result as that's what each is supposedly measuring.
Rest assured the sellers, who are in full control, have no intention of setting up rival audience measurement services, with each producing slightly different findings for each network and TV show. They know that this would produce chaos. What they want is a variety of "outcome" indicators that each seller can employ as it wishes with slected brands in addition to the standard audience projections which I am increasingly certain will come from Nielsen.
In short, there is no "problem" just a lot of confusion ---not by the sellers or the time buyers---but by almost everyone else---it would seem.
You are spot on Scott. It reminds me of my University days when my Stats. Prof. taught us a lot about what NOT to do. He used to have a saying ... "tell me the answer you want and I can deisgn the research for you ... but NEVER call your self a researcher if you EVER do that".
I have a broad question regarding US research standards. Here in AU you can't include children in research surveys under the age of 14 unless you have written specific permission from the child's parents/guardian.
For example our TV research is based on 2+ so OzTAM has to comply when constructing the panel. Radio is 10+ (it complies but the industry is comfortable with that cut-off). OOH is 14+ (again, the industry decision largely because of roadside), Readership is also 14+ (largely because kids don't purchase many newspapers). Digital is 14+ to abide by the research standards.
So, if the US standards also have 14+ limits w/o adult permission, have they considered the additional compliance cost; or 'scraping' to get under-14 data that may be accessible; or just simply define the universe as 14+ and throw-away' something like 15-20% of the TV viewing population?
John, I'm not aware of any limitation on measuring children or teens in the U.S. ---except for the difficulties in doing so and the limited advertiser interest in the results. From the very beginning, the American Research Bureau ( later called Arbitron ) included viewers under the age of 18 in its household diary studies for TV and Nielsen did the same locally and when it used diary panels to provide viewer-per-set data to be applied to its national meter set usage findings. The people meters also include children aged over one year as well as teens and the radio studies include teens. From time to time print media have funded teen redership studies as well.
I happen to agree in principle that the sellers---TV sellers in particular---are making a huge mistake sticking to hopelessly outmodd audience currency quarantees based on adults 18-49 or 25-54 as these now represent a minority of their viewers---in many cases only 25% of them. These "metrics" were never worthwhile as audience targeting devices---even when they accounted for 40-50% of the audience. But now, the sellers are not giving themselves credit for tha majority---often the vast majority---of their audience, which seems like a very poor sales approach to me. Why not use the persons 2+ "demo" or failing that adults 18+, instead and even the playing field a bit instead of giving the buyers such a huge advantage?
Thanks Ed. I'm a tad surprised that there is no limit. I know that AU, UK, NZ etc. do 'protect' kids from being bombarded. Tony may know some more.
However, having said that, the kids are already being bombarded on their 'phones and devices every day, so maybe the research industry should drop it's own standards as well.