Commentary
To Integrate Measurement -- Or Not To Integrate Measurement?
- by Maarten Albarda , Featured Contributor, April 2, 2021
I wanted to bring you some news from my tiny and very flat home country, The Netherlands. This week it was announced that a whole cluster of research acronyms (SKO, NLO, NOM, VINEX) have joined up with Kantar and Ipsos to create yet another acronym: NMO.
This really means the individual research bodies that deliver TV ratings, radio ratings, digital “ratings,” print ratings, etc., have all agreed that a combination of Kantar and Ipsos will deliver a fully integrated set of media usage data across all media. The aim is that, over time, this data set will replace all individual, siloed data sets currently in use.
There's been talk for decades about the need for an industrywide approach to audience measurement across all touch points. Here in the U.S., there have been attempts in the past, but they’ve all died sad deaths when participants could not agree on how to measure, how to share the cost, and who could participate.
In the Netherlands, we have a long tradition of compromise. Historians will tell you this is perhaps a result of having a small country that came together as one to fight off much bigger enemies like the Romans, Vikings, Spanish, French, Germans -- and, more recently, British bachelor parties. The Netherlands has a history of finding common ground across religious (we count several across our tiny country) and belief systems (anything from very strict to very liberal) for survival.
advertisement
advertisement
In the last election in March of this year, we had 37 parties in the running, of which 17 managed to win seats in parliament. The last government coalition consisted of three parties. Negotiating a compromise is in Dutch nature. There is even a globally recognized word for it: polder model.
So it’s not unexpected that all the various Dutch media industry bodies managed to get together and issue a joint RFP and jointly select suppliers. Even the choice of suppliers is a “coalition,” as it comprises both Kantar and IPSOS.
But here’s my question: Does it matter?
Up till 10 years ago, advertisers like P&G, Unilever and others pushed for better measurement across touchpoints. Today, you do not hear as much about this issue anymore. More pressing issues seem to be brand safety, diversity and audience composition and the like.
It is less about “the rating” per se and more about where audiences can be found, what drives them and how we can be sure they are even an audience (versus a bot).
It's less about delivering an audience ratings report after something has been exposed, and more about if something will have the opportunity to reach and/or move an audience at the time contact is made. Sure, you’ll need data for that, but that data is not ratings.
So does this initiative matter? And should we pursue something like it here in the U.S.?
The answer is probably “maybe.” It would be helpful to get various partners to talk and align on audience metrics, which would count for something. We would also be able to better understand relative audience contribution across a spectrum of touch points. But it feels a bit like a 1990s solution to a 2021 problem. What do you think?
Maarten, unless this consortium has found some new way to measure audiences in a meaningful as well as comparable way across media, I'm not so sure that anything has been accomplished. For example, I assume that the TV ratings ae based on something like Nielsen's peoplemeter and, perhaps there is a comparable electronic measurement for radio---or, maybe they still use diaries? As for print, I'm guessing that its a recent reading recall study for magazines and yesterday reading for newspapers.---certainly not page-specific. Last but not least, the digital ratings are, I assume, simply device usage stats based, probably, on the assumption that if an ad message appears on a user's screen for one or two seconds that that constitutes a user "impression".
Having a single source assemble all of this information and distribute it to subscribers isn't going to improve the quality of the data any and, let's be frank, when it comes to measuring advertising exposure and/or attentiveness by real people---not devices---the data is simply very inflated and poor. But not equally poor across all media types.
Also, since I assume that the individual services will continue to independently operate as before, there will be no way to calculate cross media reach and frequency patterns form a single source. Is this correct? If so, what are the primary benefits to advertisers or agency media planners/buyers and, of course, to the ad sellers? Have I missed somethig?
Ed: here is an English language press release picked from the IPSOS website.
https://www.ipsos.com/en-nl/ipsos-and-kantar-appointed-deliver-total-media-audience-measurement-solution-netherlands
From this we learn: In combining audience viewing, listening, browsing and reading through a single, integrated system it will deliver deeper insights to understand Dutch audiences and enable richer trading currencies for each media.
So you are probably right in your assessment. But... my point is that regardless whether or not this is an aggregation of some sort (which this appears to be), or a "new" methodology to measure across all touchpoints, it appears to me that either would not address the currency debate of today.
In my mind, and what I was trying to convey in my column, is that what advertisers need is a tool that can help with optimization across touch points "in the now", not after the event. Not saying an after action review is not useful or necessary, but to me I want to know how to "hit" audiences regardless of where they are. And with "hit" I want to know more than that they are merely "there" - I'd like to also know if that opportunity represents a real chance to connect and move them to a response.
Not sure I am explaining myself all that clearly. I guess I am interested in a hoslistic tool that helps me plan real time in terms or reach and effect/impact.
Maarten, I can't speak about Holland but in the U.S. estimating the combined reach of several media when more than one is included in a media plan---which, by the way, is rarely done----is a media planning, not a buying function. Here, largely for personnel cost efficiency and expertise reasons, we have separate buyers for each medium---national TV, local TV, radio, magazines and digital. These buy to specs---as best they can--- that are set up in the media plan for the venue in their area of responsibility.. They do not get involved in estimating the combined reach ot TV versus digital, print or radio buys. Indeed, they almost never get into reach at all. This has been estimated by the planners who have specified certain dayparts, program genres, channel types, etc. with, among other things, reach attainment in mind.
The clamoring for cross media comparisons is, in effect, a call to radically alter the functions of the various parties---client media director/brand manager, agency media planner and media buying specialist, in effect making those who buy planners and brand managers as well as skilled negotiaters. Moreover, a national TV time buyer would now have to be an all-media expert;a print buyer would have to know all about digital,national TV, radio,etc. LOL on that one.
The push for comparable cross media measurement is driven mainly by the digital media folks who think that, somehow, the numbers would be in their favor and this would initiate a surge of branding ad dollars away from "legacy" media and in their direction. Sorry, but that's not a viable assumption in the light of actual media selection and buying practice. What is needed is far better and less inflated measurements that are realistic for each medium. For TV it's attentive commercial audience; same for radio and digital video. But for magazines, it's probaly average ad page audience as almost all of the ads are P4C units. As for newspapers, I'll have to think about that.
Thanks Maarten for bringing this news to a broader audience. Firstly, I must admit to have had an interest in this project (I was CEO of Kantar's Media division when we bid for this contract). I would also agree that the Dutch has often been innovators in the media measurement space ...unless I am mistaken the Dutch introduced the first full online fielded national readership survey a few years ago. I must admit to being somewhat disappointed at the tone of your response to this initiative. It has taken several years to pull this together from a specification and funding perspective. Unless the specification has changed in the last 12 months, it represents the closest thing yet to the WFA (with ANA and ISBA support) to a genuine cross screen measurment.
Combining Ed's comments from this article and the one written by Joe about Politics and Media research. I am not sure that this Dutch initiative is being directly funded or at least majority funded by the big digital players but represensts a true jointly funded joint indsutry initiative. I know that there is s belief (perhaps promoted by large US research vendors?) that such joint industry funded solutions breach the Sherman anti-trust legislation). My understanding is that both the ANA and ISBA are struggling to get TV network support for their initiatives in the US and UK respectively.
As for the suggestion that this looks to be solving today's problem with yesterday's solutions. This represents a major step forward and over time may well be extended to include elements such as attention (Disclaimer-I advise TVision). Please let's celebrate this landmark for what it is...a major step forward for cross media measurement!
Hey Andy: thanks for the feedback. I agree that the initiative is worth celebrating. I felt it was an important development, which is why I wrote about it in the first place.
What I was trying to convey though, is that - while important - it feels to me as if it is solving yesterday's problem in today's world. I think that as an industry we are moving into real time audience buying across all touch points and while I do not discount the value of having a singular currency post-activity report-out, I think the true innovation needs to happen on the "looking forward" and "reporting the now" using a singular currency fashion.
Case in point, on one of my projects we are looking at a media mix of 75% digital (across all manner of platforms) and 25% linear (mostly TV and some outdoor) for the Olympics. The best "currency" on offer across all of these touchpoints is "impressions", and at a 2+ audience at that. That is disappointing to me. And for that, I wish I had a better, more granular single currency tool that can help plan and optimize "in the now" across all touch points.
Does that make sense?
I guess I should chime in again on this discussion. Unless I missed something, assembling existing survey data from various sources for various media, using metrics that apply only to each medium, is a useful first step in terms of making such information more readily available. However, it certainly does not mean that a solution to the cross platform comparability problem has been attained. I happen to believe that there probably isn't a single "metric" that presents a truly comparable advertising audience indication for TV and digital video and certainly not for print media and radio. So, until somebody finds such a metric---and documents its viability, we have a long way to go before the golden fleece is ours.
Ed: totally agree. The Dutch initiative will be insightful. But it won't solve for what you articulated so much better than I did. I applaud the step forward though, even if it is a baby step. Perhaps a crypto media currency is the answer, as crypto seems to be the answer to anything else these days :-) (PS if Crypto IS the answer, I am laying claim to that idea right here, right now! BitTRP?)