Day 5 of the asi virtual Conference, addressed Cross Media Measurement, XMM. This is one of today’s hottest media and advertising issue worldwide. Using the World Federation of Advertisers, WFA, XMM design framework requirements, referred to as the “North Star”, the session was entitled, “When you wish upon a (North) Star: will the cross-media dream come true?” The epilogue offers a straightforward pragmatic solution!
The WFA’s “North Star” initiative has spurred the development of executable solutions by ISBA (UK) with ‘Project Origin’ and the ANA (USA) in collaboration with the Coalition for Innovative Media Measurement, CIMM.
In the session’s announcement, a fundamental question was asked: “What do we really mean by ‘cross-media measurement’ in the first place?” As this final session progressed it was apparent that other, possibly better, solutions are already being developed organically at a local level from the ground up.
The session chair, Richard Asquith, noted that an XMM database could be “just a planning tool” while recognizing the value of an integrated, de-duplicated, coordinated single currency which would be used for trading. It should be noted that other industry leaders from both sides of “the pond” have also suggested it as merely a planning tool or as a media trading tool while yet others are focused on its longer-term objective to measure brand outcomes but based only on media’s contribution. Measuring just media’s contribution to outcomes without comprehensively including the complete array of marketing pressures of any campaign to the end result surely is surely a disconnect?
Richard also reminded the conference of the four cornerstone issues: funding, privacy; Joint Industry Committees, JIC, collaborations; and the different objectives for the different stakeholders. He did recognize it was a significant development for the industry on behalf of brands even though it is initially focused on TV/video platforms. Recognizing that the media fund media measurement, he considered that this initiative could destabilize currencies or disrupt media planning at a minimum. (The fundamental media adage, “Buy what you plan and plan what you buy” can surely only be achieved with a single, harmonized cross-media database?) Other critical aspects he noted: meeting local media measurement needs albeit within the North Star framework in attempting to achieve global consistency; resolution of access to walled gardens data which will be inevitably required by any solution; the influence and position of Facebook and Google which are both under scrutiny for monopolistic practices and anti-trust.
Stephan Loerke, CEO, WFA, led the day by underlining the value to the industry of this new intense advertiser involvement in ‘media measurement’ which they have traditionally left to their media agencies. The ‘North Star’ construct attempts to recognize ‘Advertisers Needs’ and ‘Industry Requirements’ which some would say reflect inherent compatibility concerns. However, the goal of real reach and frequency by target group across media vehicles would be a huge step forward although the media metric basis for R/F is currently confused. In addition, Google may be exerting undue influence on that dimension. Unfortunately, Stephan admitted that any solution for TV/video may not work for other media platforms. If this is true and the indications are that it will be, are we wasting our time? In addition, the disconnect of using XMM to describe “outcomes measurement” must be noted. As Sequent Partners, USA have essentially recommended, developing harmonized TV/video viewing or exposure data by brand target group will produce a significant improvement in the accuracy of attribution or marketing-mix modelling. It is those models and techniques that are key to the advertiser’s primary concern, understanding outcomes. The value of XMM must be recognized as a means to and end not an end in and of itself for advertisers, although it potentially offers a massive improvement for media agencies and content programmers by levelling the metrics playing field at a more meaningful level.
Julien Rosanvallon of Médiamétrie and Valérie Morrisson, CESP offered, “A perspective on the WFA cross-media initiative.” Their position was founded in the need for excruciating testing and evaluation of every element of any XMM approach and raised the concern as to whether any solution would be for planning or for buying. Their very technical assessment was founded on: the underlying assumptions of device compatibility and of independence; the sensitivity across all data elements; provision of reach & frequency estimates by provider or site due to consolidation of impressions (definition of media impressions for this initiative have still not been established!) at the very last stage plus de-duplicating TV, video and Digital (mobile) audiences; the many issues with creating virtual ID’s, VID; issue of publishers with no validated demo information; etc.
“Progress on Project Origin”, ISBA’s “local” initiative based on the WFA XMM Technical Design was shared by Richard Halton, ISBA and Joe Lewis. Richard stated that the WFA Technical Design had been, “validated, evolved and adapted” albeit that the, “TV centricity was an issue” and that “outcomes and effectiveness was to follow”. Joe reviewed the progress made in the various technical design areas including a reference to the Media Rating Council, MRC on reporting requirements and the inclusion of MRC’s misunderstood and misnamed, “duration weighted viewable impressions” which has been seriously questioned by some global media research experts. RSMB has been commissioned to develop a VID framework which will undergo a proof of concept in a prototype study and will underpin de-duplication for reach and frequency. Joe suggested that an elegant overall solution with sensible trade-offs between the accuracy and prediction of cross-media duplications versus the practicality of the process will achieved. In moving from “could” to “should” he also recognized the importance of technical infrastructure, privacy, fulfilling standards, defining metrics, embracing the role of shareholders and on-going innovation. As to next steps, ISBA will seek to broaden industry consensus, build confidence with proof of concept tests, and serve the WFA goal of global scale. Richard opined, for Project Origin, “Good enough for the UK is not enough”.
UK On-line Measurement, UKOM, appointed Ipsos to measure on-line content and ad campaign verification from 2021. Ipsos Iris’ is a hybrid methodology combining the MediaCell and RealityMine device software across a core 10,000 single source, multi-device panel of people aged 15+, with meters installed across all personal devices which passively measure the usage of websites and apps. That data is also combined with census (tag) derived data to provide the best measurement and insights possible. In, “A look into Iris – the new standard for online measurement in the UK”, Mario Paic of Ipsos and Ian Dowds of UKOM discussed the use of advanced census measurement, establishment surveys and synthetic population techniques that are part of obtaining cross device measurement – mobile, tablet, lap-top, smart phone – which overcomes the loss of cookies. UKOM believes the nature of the panel built, tagging platforms, and survey to survey integrations encompass the best of data science. The launch of the new usage service is scheduled for January 6, 2021 with ad measurement Q2.
Understandably Justin Sampson, BARB had particular, “Cross-media insights from a TAM Panel”. While BARB shares common ground with ISBA and the WFA, he underlined the need for, “consistent definitions of audience units for intersecting BARB data with any other measures.” Noticeably his slides referred to “viewing” rather than “impressions” on TV sets and other devices. He also echoed the need for duration weighted impression units as a “must have” but did not define impressions. However, he emphasized the issue of how to establish a level media playing field if impressions definitions are different in XMM as it could potentially damage the BARB currency.
As the primary funders for media measurement, the last presenters, Katty Roberfroid, egta – http://www.egta.com/ and, Robert Schäffner, IP Deutschland GmbHraised an array of concerns with, “Cross-media measurement: what’s in it for the Broadcasters?” Not surprisingly they recognized that, “there are no ‘broadcasters’ anymore. They are multi-platform”. Consequently, unbiased quality XMM is a must due to broadcasters’ evolution and there being no universal media currency available. egta has already produced an extensive insight paper, “Advances in Hybrid TV Audience Measurement” that recommends empowering local JICs, “who are already delivering multi-channel measurement and need encouragement and support”. In their opinion, JICs represent trust which is fundamental in every component of any system via co-operation and collaboration. It includes “testing, testing, testing with real data and full industry input”. Apparently, duration weighting is already accounted for and included in most European TV measurement currencies and critically established and agreed metrics must not be lost. The chasm between measures of viewability (so called “viewable impressions”) and actual viewing or exposure were noted with particular reference to calculating audience de-duplication or not! (There is no audience component with viewable impressions!) egta also emphasized the importance of brand safety, program content and context for ads. Surely factors no less important than duration? With a plea to, “not to re-invent the wheel”, egta urges that JICs are, and must be, the foundation of any XMM. Their conclusion? Do not establish a new system or structure.
Following these diverse presentations, a terrific Q&A with the speakers continued the conflicting and disconnected positions regarding whether the WFA’s technical design proposal should be oriented to media planning or trading or simply be a higher quality ancillary resource for analysis of outcomes? In the UK, the intersection between ISBA, IPA, BARB and other industry Associations on the Project Origin will be fascinating and may hold learnings for the ANA/CIMM effort in the USA.
Epilogue
Perhaps the March Hare in Alice in Wonderland has some good advice for this XMM initiative and the various Alice’s (WFA, ISBA, ANA, CIMM) involved? “Then you should say what you mean,” the March Hare went on. “I do,” Alice hastily replied; “at least — at least I mean what I say — that’s the same thing, you know.” “Not the same thing a bit!” said the Hatter. “You might just as well say that “I see what I eat” is the same thing as “I eat what I see”!”
I suggest this co-incidentally echoes exactly what Karen Nelson-Field, Amplified Intelligence wisely stated on Day 3 of the conference. “Without viewing (“seeing” or Eyes-On) there will be no chance an ad can work”. Mike Follet, Lumens, emphasized, “It is unreasonable to ask media to guarantee an outcome as they would be taking on risk that they neither control nor are accountable for.”
Interestingly Out-of-Home and radio (via PPM) already measure audience “contacts” or “hearing”, Level 3 of the ARF Media Model, in many markets around the world under the direction of the local JICs. Advertisers via JICs have ultimately always had the power to demand that all major media worldwide measure at this most meaningful, fair, and relevant level for evaluation of their media investments. So, …. moving forward could be straight forward? (The relevance of the ARF Media Model was reviewed in the review of Day 3’s session. https://archive.researchworld.com/attention-a-media-measurement-gold-standard-the-arf-media-model-has-the-answer-hint-it-is-not-attention-based/ )
As a career student of Erwin Ephron, if the WFA, ISBA and ANA “mean what they say” under the banner of Cross-Media Measurement their task is, I suggest, straightforward. Measure audience exposure or contacts across all media platforms, i.e., Level 3 in the ARF Media Model. If the industry walks away from the fundamental media/ad measurement principles, then a free-for-all will result. In that event, why would any media owner support this?
Measuring today’s media audiences with methodological quality at the true exposure level will involve sophisticated technologies and complex data integrations. However, technology should serve and enhance principles, not take their place. This is a universal business challenge not just an issue for the media and ad industries.
This asi session concluded a virtual in-depth look at Radio, TV, and Video research by an array of outstanding global industry leaders over each of 5 days that were reviewed exclusively for Research World. It was organized by Richard Marks and Mike Sainsbury of asi who are to be highly commended. This Annual asi event is expected to be “live and in -person” in a major Western European city November 2021. It is hoped that a virtual version is not the “new normal”!
