CEO & President Member Newsletters
2Q 2019: Building Consensus for Progress in Audience Measurement
At the recent Advertising Research Foundation AUDIENCExSCIENCE conference, the industry took stock of the progress made and the challenges that still lie ahead in pursuit of comprehensive and comparable cross-platform audience and advertising measurement. Though the vast majority of video viewing today can be captured accurately by existing systems, ours is always an anxious and restless industry. At AxS, some speakers voiced concern that measurement may not be keeping pace with changes in technology and in consumers’ ever-more-diverse ways of consuming media – particularly video.
However, the overall mood of the conference was more upbeat than despairing about the future of audience measurement, and in some cases the optimism was palpable. Progress was reported in coverage of all types of video consumption – whether linear, on-demand, ad-supported or subscription-based, delivered through cable or satellite or OTT. There still are some blind spots. There still are some uncooperative sellers who refuse to support the transparency afforded by independent third-party measurement, but there are fewer of them than there were at this time last year – so progress! What’s more, the independent measurement companies have gotten better at using projectable random samples and panels to calibrate and correct for the known biases in the more granular viewing data available from set-top boxes, Smart TVs and streaming video providers. In prior years, these types of hybrid solutions provoked more negative comment, but they appear to be gaining increasing acceptance as methodologically viable.
Cross-platform measurement progress is especially evident in ad campaign measurement, less so in content measurement. Here, as in many areas in our field, the barriers are not so much technical as political and driven by parochial business concerns. We know how to measure exposures across platform, but it requires the cooperation of all players in the distribution chain. If, for example, a company like Amazon would — for the common good — support unified third-party measurement, the whole industry would benefit from much more comprehensive measurement of media and advertising exposures. This is a fundamental challenge to the advertisers who pour money into companies like Amazon, but exert no pressure to cooperate in cross-platform measurement.
Here is where marketers could do more to advance their own interests, particularly in encouraging walled gardens to support open-standard SDKs and standardized identifiers such as those advanced by CIMM and the IAB Tech Lab. Advertiser pressure has already been very useful in reducing the institutional barriers to good cross-platform measurement, and more marketer encouragement would likely be beneficial. The ARF stands ready to work with other industry groups to support this goal.
Apart from coverage, progress was also evident in the evolving discussion of how and what to count (i.e., the metrics). The MRC’s George Ivie summarized their proposed new cross-platform counting standard: the duration-weighted viewable video impression. This proposed standard builds on the MRC’s prior work to develop minimum standards of viewability and filtration to eliminate non-human ad impressions (e.g. bots) – but this is the first big public effort to create a next-generation metric that bridges the gap between how linear TV and digital video are currently measured.
Regardless of how one feels about the details of the proposed standard, progress is indicated by the mere fact that the MRC has published a detailed proposal for people to debate, support and/or refine. Though most of the proposed 70-page standard enjoys widespread acceptance, George admits that 4 pages dealing with time are controversial. Some object that its time parameters are wrong – that advertising impact begins sooner or extends longer than the proposed metric allows and that, therefore, some media or delivery devices would be systematically disadvantaged were this to become a trading currency. Others argue that the time dimension of the proposed duration-weighted impression is conceptually fair, but imprecise, because ad impact is not linear – that four seconds of exposure is not twice as valuable as two seconds – just as :30 second TV ads were not twice as effective as :15 second TV ads. These debates should be welcomed – especially when they are grounded in transparent empirical research. After all, that is one of the key ways in which we make progress against difficult objectives. At AxS we saw studies addressing these questions, and more are expected in the months ahead – including one on the duration-weighting question planned by the ARF itself.
This gets to the core of the raison d’etre for the ARF. Founded in 1936, the ARF exists to apply the rigor and objectivity of scientific methods to solving key questions about how to measure and value advertising and marketing activities. It is hard to imagine a more direct way to deliver on that mission than to conduct basic research that informs the development of an industry standard that ultimately will be the basis of industry audits. Standards supported by strong empirical evidence are more likely to enjoy wide approval and legitimacy, to be perceived as fair and impartial. In this, the ARF is a natural ally of the MRC. Both organizations aspire to be scrupulously non-partisan and objective. Both see their member constituencies as including both buy-side and sell-side, and both listen closely to their members’ wishes in prioritizing activities.
This is not to say that the ARF is the “research arm” of the MRC. They are independent organizations with distinctive, if sometimes overlapping, goals. The ARF has a broader member base and thus it gets into advertising creative, branding, marketing allocation and other topics apart from media and ad measurement. The two organizations’ product portfolios are a bit different – with the MRC focused on auditing and the ARF focused on research and training – but their underlying missions are well-aligned.
Both organizations are sometimes faulted for being too slow and deliberative, too concerned with building consensus, and it is surely true that we will almost never see 100% agreement on all measurement issues. But it is equally true that measurement metrics, methods and systems that gain broad consensus and are supported by the best available evidence can have enough legitimacy to serve as the basis of business transactions. There is no substitute for legitimacy and consensus, even if they take time to build. So now is the time to act to help build consensus and move things forward. Submit your comments on the proposed standard to the MRC during its open comment period. Reach out to the ARF to see how you can get involved in its projects and initiatives to lead innovation in cross-platform measurement. And if you are a cross-platform advertiser, let your marketing partners know how you feel about the need for open, transparent and valid ad measurement – the key to managing reach and frequency and ultimately to analyzing campaign performance.
Please let me know if you have any questions.
NOTE: an abbreviated version of this perspective was reported in Mediapost [https://www.mediapost.com/publications/article/335129/building-consensus-for-progress-in-audience-measur.html]