An ad industry initiative to develop a new system enabling marketers to have sublime control over the reach and frequency of ads exposed to individual consumers without personally identifying them is expected to be operational by the U.S. as soon as late 2023, the team leading the effort said late Thursday during a panel presentation at the Association of National Advertisers (ANA) Media Conference.
“I want this to be the measurement solution that captures it all,” ANA Executive Vice President-Measurement for Marketers Nathalie Bordes said, adding, “I do not lack ambition.”
The system, which the ANA began developing in conjunction with the World Federation of Advertisers (WFA), and in consultation with the Media Rating Council (MRC), several years ago, is expected to begin a pilot utilizing live, real-world campaign data from major marketers beginning by the beginning of the fourth quarter of this year, and Bordes said the initiative expects them to test the system for “six to eight months” to ensure it is “scalable and sustainable.”
advertisement
advertisement
“That brings us into the middle to third quarter of 2023, which allows us also to plan out a productionalized solution towards the end of 2023 – early 2024,” she said.
Key components of the system are a “virtual consumer ID,” which will enable marketers, media and platforms to segment “cohort” audiences, and control the frequency with which consumers are exposed to specific ads.
The panel alluded to data cited by Procter & Gamble Chief Brand Officer Marc Pritchard during a presentation earlier in the day that estimates the average consumer is now exposed to between “6,000 and 10,000” ads daily, so part of the so-called “Cross Media Measurement” initiative’s goals it to control how, when, where and why consumers are exposed to ads to both ensure a better consumer experience, as well as improving advertising efficiency and effectiveness.
The initiative recently issued a new request for proposals (RFP) for a first-of-its-kind panel that would be used by marketers to “calibrate” their own first-party consumer data to model the reach and frequency of ads they buy in the overall media marketplace.
Importantly, Bordes said the panel will be designed to ensure “consumer consent” among all marketers, agencies, and media companies utilizing it in every country they use it in.
The panelists, which also included MRC CEO George Ivie and Procter & Gamble Global eBusiness Analytics & Insights Leader Kanishka Das, noted that the initial version of the system will focus on linear and connected TV (CTV) and online video, but that the plan is to eventually incorporate all advertising formats, as well as other forms of analytics, such as measurement of creative performance, and even so-called “outcomes-based” metrics measuring consumer behaviors after being exposed to ads.
Ivie emphasized that the MRC has been playing a strictly advisory role, but noted that it has separately been working on a variety of media measurement standards that are integral to the new cross-media measurement system, and that it would also likely end up auditing the calibration panel and any third-party measurement services utilizing it.
He described said the MRC began by creating standards for advertising exposure measurement and is close to developing new standards for outcomes measurement, and he likened the process to building a house.
“One thing you should understand, you can’t build a house without a good foundation,” he said, adding, “You need to have good exposure measurement linked with those outcomes to make a more legitimate process observable. Because if you have a certain outcome, or you don’t get a certain outcome, very often you have to go backwards and understand what happened in exposure.”
During the audience Q&A session at the end of the panel, moderate Bill Tucker, Group Executive Vice President of the ANA who has been coordinating the initiative, read a question asking if the litany of new audience measurement currencies being developed by media suppliers is compatible with the initiative, or whether it will create some confusion.
Tucker noted that the committee is “very aware of” and has been communicating with others developing and testing new audience measurement systems, but asked P&G’s Das to address it.
“The objective of the cross-media measurement initiative as envisioned by the WFA and being implemented by the ANA was never to create a new currency metric and compete with existing currencies,” he said, adding, “The objective of the cross-media measurement initiative is to restore foundational brand-building measurement on top of which we can have lots of great analytics. We can have outcomes. We can have attribution. We can do all of that. But we need to have foundational brand-building. That’s the objective.”
He ended by noting that the goals of the initiative have always been for it to be “objective, transparent, ideally third-party audited and accredited by the MRC.
“That’s a good standard for currency. That’s a good standard for metrics. Period.”
Some incredible stuff in this report, Joe. "The average consumer is now exposed to between 6,000 and 10,000 ads per day"---attributed to P&G's Marc Pritchard. Think about that. If by "exposed" we mean that the average consumer might be exposed to, say 8,000 ads per day, taking an average of the two stats cited, one wonders how such a figure was determined. For example, if we believe the TV numbers an average adult "watches" about 4 hours per day of ad-supported TV and the average ad clutter rate is around 25% when CTV is included. That works out to one hour of commercials---or, maybe, 150-160 messages, that might be seen per day. Now where do the other 7, 845 come from? Absolute nonsense.
But more important questions are how will this new system, which will----or might--- be operational in two years in many countries, be funded and, equally important, how will it measure whether anyone saw a commercial when helping advertisers to control their exposure frequencies? I guess I should include another question. What entity is going to be set up---and operational---to obtain and supply such data---The United Nations, perhaps?---and how many sellers---TV networks, stations, channels, AVOD services,etc. will buy into this plan and offer up their GRPs---oops!---I meant, "impressions", for sale and control in this manner?
@Ed Papazian: Assuming your TV advertising math is accurate, then I would say the other 7,845 come from other sources of advertising. You do know that people are exposed to ads in other places, right? I mean, there's at least one on the page you're looking at right now, so that leaves 7,844.
I haven't done any original research on this myself, but there have been observational studies that support those sort of numbers, and I think that's what Pritchard was referring to. But I'll see if we can track sources of that down.
Regarding the ad industry's Cross Media Measurement Initiative, I mainly know what the ANA has disclosed and what it's working on in the U.S., but I think the idea is a solution that the WFA and advertiser trade associations in other countries will utilize too. I think some of it might have localized solutions, but many marketers are also multinational and global, as are many media platforms, so I think geography can be semantic sometimes.
What they've been developing is a system that advertisers can use to model how virtual consumer IDs representing people in a calibration panel can be modeled as cohorts across media to control for reach and frequency.
I liked the way George Ivie described what the MRC has been doing parallel to this, in terms having a good foundation to build a house.
So the MRC started by standardizing impressions measurement. And it's now working on standardizing outcomes measurement.
Whatever those standards end up being, would be applied to the new system, and it would be up to marketers and various suppliers to figure out how to build on top of it.
And in the U.S., the MRC would continue playing the role of auditing and accrediting those solutions.
As for the rest of the world, you might be right: Maybe it will be the United Nations. But personally, I feel they have more pressing matters at hand.
Joe, I'd be facinated to see any observtional research you can dig up that finds that the average consumer is "exposed" to 6,000-10,000 ads per day. Even if by "exposed" Marc---or who ever provided that number to him----meant "might be exposed" its hard to defend such a statement. The average ad load---as a percent of content--- for all mdia is probably around 30%( just a quick estimate, Joe ) and the average consumer is counted as using roughly 9 hours per day of ad-supported media which translates into 2-3 hours of theoretical ad exposure---let's make that 2.5 hours---again just an estimate. This, of course is a vastly inflated picture as it is well documented that a very high proportion of ad "exposures" are avoided by media audiences. But sticking with the inflated figure, is it really possiblethat 2.5 hours a day of media ad "exposure" results in 6,000-10,000 ads being "exposed" per consumer per day? That works out to something like 2,400-4,000 ads exposed per hour of ad content. That's a lot of ad clutter.
@Ed Papazian: I'm not sure what your source is for 9 hours per day, but I'm looking at an eMarketer estimate that puts it at more than 13 hours daily. Additionally, when you factor in concurrent usage of media, the total amount of time the average American is exposed to media is incrementally greater than that. And that's just measured media. Regarding average ad loads being 30%, I'm also not sure where you get that from (seems like a vestige of TV and print media), but there are media where it's 100%.
In any case, I'll see if I can nail down a good observational study that puts a solid number on the total number of ads exposed daily. It looks like Pritchard was citing this article, which does not attribute a source for the 6,000-10,000 ads daily estimate:
https://ppcprotect.com/blog/strategy/how-many-ads-do-we-see-a-day/
Joe, as a guess, eMarketer is using mostly Nielsen total device usage data for digital media which counts any activity---checking email, texting, etc. as usage. I'm referring to media that carry ads---which eliminates a lot of the digital activity. But even if the 13 hour figure was correct the first question I would ask is how much of that time were ads displayed? You point out that some media content is 100% advertising. Maybe. But If I were to guesstimate the percentage for TV I would say it was around 25-30%, for audio about 25%, for Print, 50% for digital media , say 25%---in the latter case we have to make substantial discounts caused by ad blockers and fraud. So, taking all of the media I cited, together, I'd say that 30% was a fair figure and that translates into 3.9 hours of commercial exposure per consumer per day. This, in turn works out to 1,538-2,564 ads "exposed" per ad content hour per day per consumer---clearly a wildly inflated number.
I'll go even beyond that and point out that in the real world a huge portion of ads that are "exposed" ---using the might have seen/read/listened to definition---- are totally avoided by a number of means which I'm sure you are familiar. So that gigantic---and improbable---- figure of 3.9 hours per day---which probably includes multitasking --- of ad content exposure needs to be cut way down to about 1.5 hours per day. I wont even calculate how many exposed ads per hour that comes to. Sorry. It's not the biggest issue in the media world but the 6,000-10,000 figure is nonsense. I'm sure it was used to dramatize the other points that were being made and the other questions I raised are far more important. However I think that puffing things up to grandiose proportions as this seemingly factual statement does needs some push back.
@Ed Papazian: I hate to break it to you, but digital media -- including email, texting, etc. -- also are forms of advertising.
https://www.mediapost.com/publications/email-marketing-daily/
I think some of your incredulity may have to do with what you define as advertising. If feels to me like you're only considering major "ad-supported media" and factoring based on their ad/content ratios. That's certainly one way of thinking about it.
Another way, is to look at the total number of ads people are actually exposed to during their day.
In the late 1970s, Gabe Samuels did a study like that when he was media research chief at J. Walter Thompson, and they estimated it was 500 ad exposures daily.
Over the years, I've seen it grow to estimates as high as 12,000, so the numbers Marc Pritchard did not shock me.
Joe, I doubt that the numbers in question actually came from Marc. More likely one of his media research people. As for your finding the figures to be more or less OK, consider what they would mean if true. Ask youself how many hours a day is an average person exposed to ad-supported media. Is it 13 as you say eMarketer puts it? Let's agree, for the sake of this little debate, that 13 hours a day is a correct estimate. Now I ask how much of that 13 hours consists of ad messages? Is it 100%---or all 13 hours? Which would mean that the average consumer consumes nothing but ads when he/she uses media---no programs, no editorial matter, no music, etc. ---only ads. OK---so far so good. Now, let's take the mid point of Marc's range and assume that the average consumer is "exposed" to 8,000 ads per day. That means that during an average hour of ad consumption the typical consumer is "exposed" to 615 distinct ad mesages. Even for a medium that offers ad content exclusiveley that's a pretty heavy dosage wouldn't you agree? And, one might also ask, whether our mythical average consumer would really devote 13 hours per day to media which only displayed or provided ads? Do people like ads that much?
Every year we at Media Dynamics Inc put out independent estimates ---based on a review of lots of time spent, ad clutter and attentiveness data---that describe the average consumer's exposure and intake of ad messages. These are only estimates---but we feel that they are reasonable ones. And the plain fact seems to be that the average consumer notes and absorbes only a fraction of the ads that are presented by the media he/she uses. It gets even worse when you consider the amount of time devoted to ads ----even when noted. It's far smaller than many people believe. When people cite wildly puffed up stats in order to hype an article, report or speech---and rest assured, the "P&G" figures will now be cited by lots of folks as "facts" ---someone has to pushback and make people think things out. Do such statements---no matter who makes them---myself included---make sense?
@Ed Papazian: I provided a link to the source Marc Pritchard cited in my earlier reply. Discern what you want from it.
So far, you still haven't provided any sources for your estimates, so I'll infer they are Media Dynamics'.
As I said previously, it all comes down to how you define things, and I think you're thinking linearly and not considering the reality of concurrent use of media.
Betsy Frank did some mind-blowing research on that back in the MTV days, which I wish I had access to now. And if anything, it's only grown with the availability of new internet-connected devices.
But I did come across a study she worked on for the Center for Research Excellence, and which estimated it is nearly 15 hours a day when you add it all up. It's probably grown since then:
http://www.researchexcellence.com/files/pdf/2016-04/id336_concurrent_platform_usage_study.pdf
Joe, if 13 or 15 hours a day was a legitimate average for consumption of ad-supported media---which is way over the top of reality in my view----that would mean that about 20% of the population---the heaviest media consumer quintile---averages 25-30 hours per day with ad-supported media and,within that the top decile probably devotes 30-40 hours per day to such activities----unless we think that the average usage --- 13 or 15 hours ---applies to everybody. That doesn't leave much time for sleeping, bathing, going to work , etc. for most media users and especially the heavy users, does it? Sorry, there's no way one can rationalize the "P&G" figures. And as I keep pointing out---regardless of the definition employed and the nature of the research supposedly supporting it,that in the real world nothing appraoching such a mind boggling ad consumption---or "exposure"--- rate exists. If it did we would all be walking around like zombies---our minds blown away. Also, you'd have to revise your ad spend tallies---like increase them ten -fold---to account for the cost of all of those non-exostant "impressions"---unless 90% of them are makegoods.
Regarding my estimates, at Media Dynamics we examine many surveys and sources but come up with our own estimates---which have proven quite accurate---though not 100% perfect in all cases. When we see compelling new info that suggests a need to rethink or revise our estimates we do so. I should also note that the figures I cited are not our official estimates---only close approximations for the sake of this discussion.
Joe & Ed, just be reloeved that it will all be solved by 2024. After all they will be testing the system for six to eight months. What could possibly go wrong?
I agree that TV is just one (expensive) channel which runs a lower ad frequency. But media such as OOH are 24/7. Ads are not 'inserted'. The media is the ad.
Having said that I find that the the 6k-10k daily is hard to swallow. There are 86,400 seconds per day. I'll use 8,640 ads as the mid-point, which means that on average every person sees an ad every 10 seconds. I find that hard to acept as reflecting reality - maybe I like my sleep more than others. One could also say that the research behind the quoted claim might need some testing for validation.
The 'ads per day' claims seemed to escalate when broadband delivered the requisite speed. Every page/screen seems to carry at least one ad which I suspect has ramped up the ads per day. As Ed reiterates, an ad 'served' does not equate to an 'ad seen' or even an 'ad with attention'. A good starting point would be to 'discount' traffic counts and robust data on that alone would help the cause.
This marvellous dialogue reveals, once again, the fundmantal issue with media terminology as used today. The lack of precision has lead to the misunderstanding, and often deliberate abuse, of key media terms, notably "impressions" and "exposed" (per Marc Prichard) and whether such terms are even based on a legitimate device, or distinctly, a persons measure.
As John Grono eloquently pointed out, which Ed & I have consistently supported, an ad served (or a so called and misnamed, "viewable impression" which should be referred to as "content rendered counts") does not equate to a tradtionally defined impression, or opportunity-to-see by a target group of persons; which does not equate to a liklihood-to-see or LTS impression by a target group of persons; which does not equate to a contact (Eyes-On or Ears-On impression); which does not equate to "attention" by a target group of persons - primarily a creative measure!
As Lumens and TVision have consistently found, measures of actual contacts (a media measure) and more signiificantly attention (a creative measure) by target groups are wildly different and significantly lower than device measures - "viewable impressions". They also confirm that it is only ads that generate attention which can expect to achieve a brand outcome.
So, when using the terms "impressions" or "exposed" please precisely define them and verify their foundation. It may help understand, and also help dismiss, some of the numbers being thrown around and protect some of our esteemed leaders from sprouting unadulterated twaddle. Perhaps embracing this precision, will help the industry finally understand the valid foundations of media measurement versus creative measurement versus outcomes measurement and their complex relationships???